Assessing Regional Public Transport Evaluation Indicators for Advanced Mobility Systems: A Review

Abstract

The emergence of advanced mobility systems, including Demand Responsive Transport (DRT), shared mobility, and Mobility as a Service (MaaS), has required a reassessment of the evaluation indicators for public transportation systems. Existing studies often address only limited aspects and lack a comprehensive, structured classification, while the unique impacts of advanced systems remain insufficiently captured. Moreover, little attention has been given to which indicators are suitable for simulation despite their growing role in transport planning. To fill these gaps, this study develops a structured classification of quantitative evaluation indicators from the existing literature, serving as a foundation for assessing advanced mobility systems. It highlights system-specific characteristics, identifies relevant indicators, and examines their correspondence with conventional ones. Furthermore, it explores the applicability of these indicators in simulation environments, offering guidance for selecting representative indicators in simulation setup, operational monitoring, and impact assessment. Finally, it highlights the potential of quantitative indicators to approximate qualitative ones, suggesting directions for future research in simulation-based evaluation. By integrating environmental, economic, and societal dimensions, this study contributes to a sustainability-oriented framework for evaluating advanced mobility systems, providing insights for both academic research and practical mobility planning.

1. Introduction

The rapid evolution of regional development and advancements in technology are driving significant changes in public transport systems. In particular, regional public transportation faces growing challenges due to increasing reliance on private automobiles [1], leading to issues such as traffic congestion [2], environmental degradation [3], and social exclusion [4]. To alleviate these issues, various advanced mobility systems, such as Demand Responsive Transport (DRT) [5,6], shared mobility [7], micromobility [8], autonomous mobility [9], Personal Rapid Transit (PRT) [10], and Mobility as a Service (MaaS) [11], have emerged. These systems offer enhanced flexibility, efficiency, and user-centric features, advancing the capabilities of traditional public transit.

When regions plan to introduce advanced mobility systems or modify existing public transportation, pre-implementation evaluations are critical. These evaluations are often conducted through experiments and surveys to collect both quantitative data and qualitative feedback. While these methods provide comprehensive insights, they are often time-consuming, resource-intensive, and may lack scalability for long-term assessments [12,13]. Simulation, as an alternative, offers significant advantages in terms of cost-effectiveness and applicability across various evaluation scenarios [14], although it is limited in handling qualitative aspects. To enhance the future applicability of simulation in public transport evaluation, this study focuses on identifying and analyzing quantitative indicators used in public transport.

Although previous research has extensively studied specific aspects of public transport evaluation, such as reliability [15] and accessibility [16], a comprehensive classification of evaluation indicators remains lacking. Even studies incorporating multiple indicators often focus on input–output efficiency models (e.g., data envelopment analysis, stochastic frontier analysis) without developing a systematic framework for classifying indicators [17]. Beyond these approaches, which evaluate efficiency after input and output variables are defined, this study provides a structured classification to support researchers and policymakers in identifying and organizing such variables in the first place. In this way, it not only complements DEA and SFA but also enables a more comprehensive perspective that integrates frontier-oriented efficiency views with broader planning and impact considerations. This framework is essential for guiding decision-making and for simulation-based evaluation, where a clear mapping of indicators to input and output variables is critical. This study systematically categorizes public transport evaluation indicators, distinguishing between direct and indirect measures. Direct indicators are further divided into planning and operating phases, each incorporating supply-side and demand-side assessments, while indirect indicators capture broader economic, environmental, and societal impacts.

The impact of certain advanced mobility systems on public transport remains uncertain. For example, shared mobility can either complement or compete with conventional transit, influencing sustainability outcomes in complex ways [18]. Similarly, while MaaS aims to integrate multiple modes, its long-term effects on public transport ridership and system efficiency are still unclear [11]. Existing evaluation frameworks, originally developed for traditional fixed-route systems, may not fully capture these evolving dynamics. This study discusses the specific evaluation needs of advanced mobility systems, identifies key indicators tailored to these needs, and explores their relationship with existing indicators.

Furthermore, as simulation-based evaluation becomes increasingly important in transport planning [19], existing studies have not systematically explored which evaluation indicators are suitable for simulation environments. This study explores the applicability of classified indicators within simulation environments, providing guidance on indicator selection for simulation setup, operational monitoring, and broader impact assessment.

This paper is positioned as a theoretical exploration and framework development. It synthesizes and organizes quantitative indicators and discusses their applicability to advanced mobility and simulation settings; it does not provide empirical validation.

This study has three main objectives: first, to consolidate and categorize quantitative evaluation indicators currently used in public transport systems; second, to discuss their applicability for evaluating advanced mobility systems in various evaluation contexts; and third, to explore the integration of these indicators into simulation environments. By providing a structured analysis of these indicators, this paper has the potential to contribute to the development of more effective simulation tools that can support the planning and evaluation of advanced mobility systems.

The remainder of this paper is organized as follows: Section 2 describes the methodology for selecting and reviewing the indicators. Section 3 categorizes these indicators into direct and indirect evaluations. Section 4 discusses the evaluation focus for advanced mobility systems and the selection of indicators for simulation environments. Finally, Section 5 concludes the paper with a summary of findings and suggestions for future research.

2. Review Method

This study adheres to a structured methodology for reviewing the literature on the traditional evaluation of regional public transport, guided by the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) framework [20]. The methodology includes a database and search strategy, a screening process, and the establishment of inclusion and exclusion criteria.

2.1. Database and Search Strategy

The literature search was conducted using the Scopus database, selected for its extensive collection of peer-reviewed journal articles. To maintain consistency, the search was limited to articles published in English. Keywords were carefully chosen to capture the scope of public transport evaluation, incorporating terms like “public transport” and “evaluation” and specific modes of transport such as “bus”, “taxi”, “rail”, and more.

The specific search query used was (TITLE-ABS-KEY ((public AND transport) OR (public AND transportation) OR (public AND transit)) AND TITLE-ABS-KEY (evaluation OR assessment OR measurement) AND TITLE-ABS-KEY (bus OR taxi OR rail OR subway OR bicycle OR bike)) AND (LIMIT-TO (SRCTYPE, “j”)) AND (LIMIT-TO (DOCTYPE, “ar”)) AND (LIMIT-TO (LANGUAGE, “English”)).

2.2. Screening Process

To efficiently manage the large number of retrieved articles, ASReview software (version 1.4) [21] was employed for the screening process. ASReview implements an active-learning loop: researchers first manually label a small subset of records as relevant or irrelevant; the system trains a classifier on these labels and ranks the remaining records by predicted relevance. During the screening, each new label updates the model and the ranking so that likely relevant studies are reviewed earlier. Two researchers (RD and FK) screened independently. Their results were then compared, and records with different labels were discussed until a consensus decision was reached. The screening process for each researcher was stopped when ten consecutive records were labeled as irrelevant.

2.3. Inclusion and Exclusion Criteria

Articles were excluded if they lacked a Digital Object Identifier (DOI), ensuring both credibility and accessibility of the sources. The geographical scope was confined to studies relevant to regional transport, excluding those that focused solely on national-level transport. Additionally, articles centered on infrastructure risk assessments, Environmental Impact Assessments (EIA) of bus types, or evaluations related to air and sea transport were excluded, as they fell outside the scope of this study.

Moreover, studies that focused on factors affecting specific indicators—such as research exploring the determinants of user satisfaction (e.g., comfort, cleanliness)—were not considered, as the review aims to assess broader system performance rather than individual influencing factors. Similarly, theoretical studies without real-world examples were excluded to ensure the focus remained on practical, applicable findings. In cases where particular indicators, such as accessibility and equity, were heavily represented in the literature, a representative selection of similar studies was included rather than attempting to exhaustively review all papers on those topics.

3. Review Results

3.1. Search Results

As shown in Figure 1, a total of 3645 records were initially identified through database searches. After removing duplicates, 3643 records remained for screening. Title screening narrowed the selection, excluding 3480 records as irrelevant. The abstracts of the remaining 163 records were reviewed, with 106 further exclusions due to irrelevance. This left 57 records for a full-text eligibility assessment, of which 43 were deemed relevant. Additionally, 32 relevant papers were identified through references from other sources (e.g., citation searching, manual search, personal knowledge), bringing the final count of articles included in the review to 75.

3.2. Categorization of Indicators

Given that each reviewed article often addressed multiple evaluation indicators, covering various aspects of public transport systems, we first categorized these indicators systematically, facilitating a better understanding of how they are applied to evaluate different dimensions of public transport.

The indicators identified in the review were categorized into two main groups: direct evaluation and indirect evaluation. Direct evaluation indicators measure the performance of the public transport service itself, while indirect evaluation indicators assess the broader impacts of the service on the environment, economy, and society.

Within the direct evaluation category, indicators are further divided into supply-related and demand-related evaluations. Supply-related indicators focus on the provision of services, such as network design and resource allocation, while demand-related indicators assess how well the system meets user needs. Both supply and demand evaluations are further divided into two perspectives: planning and operating. Planning indicators focus on how the public transport system is designed, including strategy formulation and long-term development. In contrast, operating indicators evaluate the daily management and delivery of services, ensuring the system runs efficiently in practice.

It should also be noted that some indicators intentionally appear in more than one category because the same measure can have different meanings depending on the evaluation perspective. For example, travel time under the demand–operating category represents the minutes experienced by passengers and directly influences their satisfaction, whereas under the supply–operating category it reflects the vehicle’s running time, affecting scheduling efficiency and operating cost. Even if the same physical time is observed, it carries different meanings for users and operators. In utility or economic evaluation, user time can be translated into disutility or monetary value through value of time, whereas operator time is usually represented as operating costs such as wages and fuel. For such dual-perspective indicators, treating them separately ensures that each is assessed within its own dimension, which naturally avoids double counting while maintaining comprehensiveness.

Direct Evaluation Indicators:

• Supply–planning indicators focusing on network design, resource allocation, financial investment, and service planning to establish a solid foundation for operations.
• Supply–operating indicators focusing on service output, time performance, operational speed, and costs to ensure smooth and efficient operations.
• Demand–planning indicators focusing on understanding and quantifying transport needs, such as population distribution, land-use patterns, and demand coverage.
• Demand–operating indicators focusing on the evaluation of how transport services are utilized in real-world conditions and experienced by passengers during operation, such as ridership, travel time, and safety.

Indirect Evaluation Indicators:

• Environmental impact indicators focusing on how the system affects the natural environment.
• Economic impact indicators focusing on how the system contributes to regional economic development.
• Societal impact indicators focusing on how the system promotes social inclusion and benefits communities.

3.3. Direct Service Evaluation

The following sections detail the indicators for direct service evaluation, categorized into supply–planning, supply–operating, demand–planning, and demand–operating.

3.3.1. Supply–Planning

Supply–planning indicators focus on network design, resource allocation, financial investment, and service planning to establish a solid foundation for operations. The indicators are organized into categories, with each category containing specific definitions. The classification of indicators for supply–planning is shown in

Table 1. Classification of indicators for supply–planning.

Category	Definition	Indicator
Network Design	Network design evaluates the physical characteristics and structural layout of a public transport network, focusing on key indicators such as stop density, network density, and network length. These attributes define the spatial structure and overall extent of the network. Additionally, the design considers connectivity and centrality to assess how effectively nodes are linked and the relative importance of individual network elements. Indicators like route directness, the number of intersections, and route overlap reflect the network’s ability to support efficient travel and minimize redundancy.	stop density network density connectivity number of routes centrality network length route directness number of stops coverage route overlap number of intersections
Resource	Resource refers to the physical assets and human resources that support the operation of public transport services. This includes factors such as fleet size, fleet features, and fleet age, which impact the availability and quality of services. Human resources, represented by the number of employees, are essential for maintaining and operating the service effectively.	fleet size fleet features number of employees fleet age
Finances	Finances refer to the financial resources and expenditure necessary to support the operation, maintenance, and development of public transport services. This includes aspects such as operational costs, capital costs, total costs, and subsidies, which together ensure the system’s financial sustainability.	operational cost subsidy capital cost total cost
Service Provision	Service provision refers to the amount of service planned to meet customer needs. This category encompasses indicators such as service frequency and operating time, which collectively determine the availability of the transport system.	frequency operating time

.

The specific descriptions of each indicator are as follows.

For Network Design:

• Stop density: This indicator measures the number of stops or stations per line or area. [22,23,24,25,26,27,28,29,30,31] and the average length between two stops or stations [32,33].
• Network density: This indicator primarily refers to the total length of the public transport network within a given area, measured in kilometers per squared kilometer [29,34,35,36,37]. It includes related indicators such as the proportion of the road network served by public transit [34,38].
• Connectivity: This indicator refers to the ability of a transit network to link nodes efficiently across routes, schedules, and spatial activity patterns [39,40]. It includes related indicators such as the transfer availability based on walking time and distance [41], the ratio of stops with transfer opportunities within 500 m [42], the number of edges in routes between nodes, and the clustering coefficient [43].
• Number of routes: This indicator measures the total number of transit routes within the network [24,44] or those serving a specific area [28,29,31,45].
• Centrality: Evaluates the importance of individual nodes within a network, based on their connections, position, and influence on overall network flow. For example, degree centrality reflects the number of direct connections [43], while betweenness centrality measures how often a node lies on the shortest paths [43,46,47]. Similarly, closeness centrality assesses proximity to all other nodes [46], and eigenvector centrality considers connections to highly influential nodes [46].
• Network length: This indicator primarily refers to the total length of the public transport network, measured in kilometers [27,32,36,48]. It includes related indicators such as the network length growth rate [32].
• Route directness: This indicator measures how closely a transit route follows the shortest possible path. It is assessed by deviations from the shortest network distance [22,27] and the straight-line distance [23].
• Number of stops: This indicator counts the total number of stops or stations within the transit network [32,36,49].
• Coverage: This indicator indicates the proportion of the area that is accessible within a set distance (e.g., 400 m) from transit stations [28,34].
• Route overlap: This indicator measures the proportion of one route’s length that overlaps with another, indicating the extent to which two routes coincide geographically [41].
• Number of intersections: This indicator counts the total number of intersections on a bus route [44].

For Resource:

• Fleet size represents the total number of vehicles in the public transport system [5,25,32,34,36,48,50,51,52,53].
• Fleet feature describes the characteristics of vehicles, including E-ticketing, safety measures, on-board information, climate control systems, and accessible features [27,30,33,36,54,55].
• Number of employees indicates the workforce size responsible for operating and maintaining the transport system [48,50,56,57].
• Fleet age refers to the average age of vehicles in the fleet [36,54,57].

For Finances:

• Operational cost refers to projected expenses associated with maintaining a specified fleet size and includes depreciation, cost per vehicle-kilometer, and cost per vehicle-hour [9,30,32,56,58,59].
• Subsidy denotes the percentage of income or funds allocated to public transport, including total government subsidies and the share of transport subsidies directed to remote areas [36,48,50,55].
• Capital cost represents the expenses related to the construction and development of transit systems, such as infrastructure and vehicle investments [60,61].
• Total cost combines various costs, including capital, maintenance, and operational expenses [61].

For Service Provision:

• Frequency refers to the planned number of transit services scheduled within a specific time frame, such as hourly or daily [26,35,41].
• Operating time indicates the planned span of service hours for each route in a working day [26,37].

3.3.2. Supply–Operating

Supply–operating indicators focus on service output, time performance, operational speed, and costs to ensure smooth and efficient operations. The indicators are organized into categories, with each category containing specific definitions. The classification of indicators for supply–operating is shown in

Table 2. Classification of indicators for supply–operating.

Category	Definition	Indicator
Service Provision	Service provision refers to the actual output of services provided during the operation phase, including indicators that reflect the extent and scale of active services. Key indicators, such as operating length, frequency, operating time, active fleet size, and passenger capacity, measure the volume and accessibility of services made available to meet travel demand.	operating length frequency operating time active fleet size passenger capacity
Time	Time refers to the temporal aspect of the transport system during operation, focusing on journey duration and scheduling. Indicators include headway, trip time, running time, dwell time, and layover time, all of which describe the allocation and management of time in transport operations.	headway trip time running time dwell time layover time
Reliability	Reliability refers to consistency and dependability of the transport system during operation. Indicators include punctuality, regularity, delay, and disruption, which reflect the system’s ability to provide a reliable service.	punctuality regularity delay disruption
Speed	Speed refers to the actual operational speed indicators achieved during service, reflecting the performance of the transport network in real-time conditions. This includes commercial speed, running speed, and peak-hour speed, as well as acceleration and deceleration, which together indicate the effectiveness of service delivery, time management, and ride quality.	commercial speed running speed peak-hour speed acceleration deceleration
Finances	Finances in the operating phase focus on the ongoing costs and expenses associated with daily operations, ensuring the financial sustainability of the service. Indicators such as operational cost and incident cost measure the financial efficiency of the system, reflecting the economic demands of sustaining reliable public transport.	operational cost incident cost

.

The specific descriptions of each indicator are as follows.

For Service Provision:

• Operating length refers to the actual total distance covered by transit services during a working day [5,29,32,36,44,48,53,54,57,62,63,64].
• Frequency measures the actual number of transit services provided within a specific time frame [22,27,28,30,31,33,51,64].
• Operating time indicates the actual span of service hours for each route during the day [22,27,33,44,53,62,63].
• Active fleet size represents the number of vehicles actively operating within the network at any given time [5,7,25,32,53,57,59].
• Passenger capacity refers to the total number of passengers that the transit system can accommodate within specified timeframes, such as per hour, per day, or per month [5,23,27,49].

Although supply–planning also includes service provision indicators, these are focused on anticipated service levels during the planning phase. In contrast, supply–operating’s service provision indicators reflect the actual services provided, serving as a monitoring tool to assess real-time performance. Most studies rely on operating-phase indicators, as they provide a more accurate representation of the system’s effectiveness and efficiency in meeting passenger demand.

For Time:

• Headway refers to the time interval between two consecutive vehicles on the same route [37,42,44,61].
• Trip time refers to the total time that elapses during a complete journey for a service vehicle from the original terminal stop to the last stop along the route [25,26,36,44,58].
• Running time refers to the actual time taken by a vehicle to travel between two consecutive stops during operation. This excludes time spent at stops for boarding and alighting, as well as waiting time due to holding controls or other operational delays [47,58].
• Dwell time refers to the service time to allow passenger boarding and alighting activities and possibly additional time due to holding control [58].
• Layover time refers to the scheduled time interval at a terminal or stop between consecutive trips to allow the transport service to recover from delays and variability in trip times. It is measured as the difference between the average planned trip time and α-percentile of the vehicle total trip time distribution across both directions of a route [58].

For Reliability:

• Punctuality measures the on-time performance of transit services, indicating the degree to which services adhere to their scheduled times [33,34,42,44,48,61,65,66,67].
• Regularity refers to the ability of the transport system to maintain consistent intervals between successive vehicles at a given stop or along a route, often measured by the coefficient of variation (COV) of headways [58,65,66].
• Delay measures the extent to which transit services deviate from their scheduled times, covering indicators such as the number of delay events, average delay per vehicle per kilometer, and the difference in travel times during peak and off-peak periods [25,32,38,42,68].
• Disruption refers to the total number of passengers who did not complete their journey on the urban rail transit due to excessive wait times or changes in the network [68].

For Speed:

• Commercial speed represents the average speed of transit vehicles, including both running time and stop time, over a complete route [27,30,36,37,41,42,61,69].
• Running speed refers to the speed of the transit vehicle while it is in motion, excluding any stops [32,33,42].
• Peak-hour speed measures the average speed of transit vehicles during peak travel hours [34,48,61].
• Acceleration describes the variability of transit vehicle acceleration during operation, measured by the variance of vehicle acceleration [69].
• Deceleration refers to the frequency and intensity of braking events, commonly quantified as the number of vehicle decelerations exceeding a threshold (e.g., 3 m/s²) [69].

For Finances:

• Operational cost encompasses various ongoing expenses associated with daily operations, including fuel/power, maintenance, and wages [6,11,30,32,36,60].
• Incidents cost reflects the financial impact of unexpected events, such as accidents or collisions [30,60].

In the operating phase, these costs are monitored as part of routine management, including expenses for unforeseen situations, distinguishing them from planned costs in the planning phase.

3.3.3. Demand–Planning

Demand–planning indicators focus on understanding and quantifying transport needs, such as population distribution, land-use patterns, and demand coverage. The indicators are organized into categories, with each category containing specific definitions. The classification of indicators for demand–planning is shown in

Table 3. Classification of indicators for demand–planning.

Category	Definition	Indicator
Socioeconomic Information	Socioeconomic information refers to the population characteristics and spatial patterns that influence the demand for public transport. This category includes indicators such as population and land use, which together provide insights into potential user bases and population distribution.	land use population
Potential Travel Needs	Potential travel needs quantify potential demand for transport services based on demand coverage and specific user groups.	demand coverage user-group characteristics

.

The specific descriptions of each indicator are as follows.

For Socioeconomic Information:

• Population refers to the population and socioeconomic attributes of the area, providing insights into user bases and income-level disparities [29,40,48,60].
• Land use reflects the distribution and diversity of land types within the transit service area, including residential, commercial, official, and public service spaces [29,38,40].

For Potential travel needs:

• Demand coverage refers to the proportion of potential users within a defined service area that is effectively served by the public transport system. It evaluates the spatial alignment of transport services with population distribution [38,42].
• User-group characteristics refer to the attributes or behaviors of specific passenger groups, such as cross-border travelers, providing insights into their unique needs to support targeted transport planning [38].

3.3.4. Demand–Operating

Demand–operating indicators focus on the evaluation of how transport services are utilized in real-world conditions and experienced by passengers during operation, such as ridership, travel time, and safety. The indicators are organized into categories, with each category containing specific definitions. The classification of indicators for demand–operating is shown in

Table 4. Classification of indicators for demand–operating.

Category	Definition	Indicator
Service Utilization	Service utilization refers to the extent of service usage by passengers during operation. This includes indicators such as ridership, load factor, and number of trips, which collectively indicate the level of demand and how well the service capacity aligns with actual usage. These indicators help assess the system’s ability to meet user needs and optimize resource allocation.	ridership load factor number of trips passenger-kilometer mode share travel distance complaints mode choice probability active hours empty distance trip completion rate
Time	Time encompasses the temporal aspects of the passenger experience, measuring various components of travel duration. Indicators such as wait time, travel time, in-vehicle time, and walking time reflect the efficiency and convenience of the transit service from a user perspective. These indicators are crucial for evaluating the overall service quality and minimizing passenger delays.	wait time travel time in-vehicle time out-of-vehicle time time saving
Finances	Finances in the demand–operating context refer to user-related costs and revenue generated from service usage. Key indicators include fare cost and revenue, providing insight into the financial sustainability of the service and its affordability for users.	revenue fare
Transfer	Transfer refers to the demand for and efficiency of passenger transfers within the transit network. This category includes indicators such as transfer time, transfer rate, and transfer ridership. These indicators assess both the convenience of transfer options and the actual need for passengers to switch services, which is essential for understanding multimodal connectivity and optimizing transit flow.	transfer time number of transfers transfer ridership transfer rate
Safety	Safety refers to the risk factors affecting passengers during transit, including indicators such as number of accidents, accident frequency, and crime incidents. This category assesses the reliability of the transit system in protecting users from harm and ensuring a secure travel environment.	number of accidents accident frequency number of casualties number of crimes

.

The specific descriptions of each indicator are as follows.

For Service Utilization:

• Ridership refers to the total or average number of passengers using the transit service over a specific period, such as daily, monthly, or annual counts. Ridership also encompasses growth rates, turnover volumes, and passenger distribution across routes and stations [5,22,25,29,30,32,36,42,47,48,49,51,52,53,62,70,71].
• Load factor measures the utilization of seating and vehicle capacity, including the ratio of passengers to available seats, occupancy rates during peak hours, and overall seat utilization per run or per month [5,27,30,33,34,37,42,45,51,57,62,68,70,72,73].
• Number of trips represents the total and average quantity of passenger trips within the transit system, including daily trips, trips per area unit, and trips per population ratio. This indicator also encompasses growth rates, trips by passenger category, trips per vehicle-kilometer, and trips per revenue hour [34,38,54,60,61,69,74].
• Passenger-kilometer measures the total distance traveled by all passengers within the transit system, including indicators such as total passenger-kilometers, passenger-kilometers per capita, and the ratio of passenger-kilometers to vehicle-kilometers [36,38,44,59,60,62,72,75,76].
• Mode share represents the proportion of total trips made by a specific transportation mode, reflecting the relative usage of that mode within the overall travel demand [34,36,55,59,61,71].
• Travel Distance: Represents the average distance traveled by each passenger within the transit network, reflecting typical user trip lengths. This indicator is distinct from the passenger-kilometer, which focuses on the total travel distance for all passengers, as travel distance emphasizes the range of individual trips [32,36,69,71].
• Complaints tracks the number of passenger complaints received and effective passenger complaint response rates [32,33].
• Mode choice probability measures the likelihood of passengers choosing the transit service over other transport options [77].
• Active hours refers to the total number of hours the service is actively used by passengers [9].
• Empty distance represents the distance traveled by vehicles without passengers during service operations [9].
• Trip completion rate represents the percentage of passenger trips that successfully reach their destinations within a specific operational period [72].

For Time:

• Wait time measures the average time passengers spend waiting for a transit service [23,35,51,52,58,66,72,78].
• Travel time represents the total time taken by passengers from the point of origin to the destination, including all stages of the trip [11,23,47,58,71,72,78,79].
• In-vehicle time reflects the time passengers spend inside the transit vehicle during their journey [23,34,52,58,78].
• Out-of-vehicle time indicates the time passengers spend outside the vehicle during a trip, such as during transfers [35,78].
• Time saving refers to the reduction in travel time achieved through advanced mobility systems, such as ride-sharing or bike-sharing, compared to baseline scenarios like traditional public transport modes or non-integrated trips [73,77].
• Walking time measures the time passengers spend walking to and from transit stops [33,35].

While time indicators are also part of the supply–operating phase, there they focus on management aspects, such as schedule adherence and service intervals, assessing the system’s operational efficiency. In the demand–operating phase, time indicators instead capture the passenger experience, focusing on convenience and service quality from the user’s perspective.

For Finances:

• Revenue encompasses various income streams generated from passenger fares and ticket sales, measured through different indicators, such as total revenue per vehicle, per operator, or per kilometer traveled. Additional indicators, like fare recovery ratio, offer insights into financial efficiency by comparing revenue against operational costs [6,30,32,36,38,48,50,51,56,57,62,68].
• Fare reflects the price paid by passengers to use the transit service, providing insight into affordability based on pricing [23,33,36,37].

In the demand–operating context, these financial indicators focus on user-related costs and revenue generated directly from service usage, offering a view into the transit system’s financial sustainability and user affordability.

For Transfer:

• Transfer time measures the time passengers spend switching between transit services [23,24,42,45,78].
• Number of transfers reflects the total number of transfers made by passengers throughout their journeys [23,24,26].
• Transfer ridership reflects the number of passengers who make transfers, indicating the demand for transfer facilities [32,45].
• Transfer rate represents the proportion of trips involving transfers within the transit network [24,30].

For Safety:

• Number of accidents represents the total count of accidents occurring within the transit network over a specific period [30,32,33,40,69].
• Accident frequency measures the rate of accidents per operational unit, such as per kilometer or per service hour [32,54,57].
• Number of casualties indicates the number of injuries or fatalities resulting from accidents within the transit system [38,54,55,69].
• Number of crimes reflects the recorded instances of criminal activities [40,69,80].

3.4. Indirect Service Evaluation

The following sections detail the indicators for indirect service evaluation, categorized into environmental impact, economic impact, and societal impact. This categorization aligns with the widely accepted framework for sustainable development, which emphasizes balancing the needs of present and future generations in environmental, economic, and social aspects [81]. These three dimensions are particularly relevant to public transport and advanced mobility systems, as they encompass the key areas of sustainability—minimizing environmental externalities, supporting economic growth, and enhancing social well-being.

3.4.1. Environmental Impact

Environmental impact indicators focus on how the system affects the natural environment. The indicators are organized into categories, with each category containing specific definitions. The classification of indicators for environmental impact is shown in

Table 5. Classification of indicators for environmental impact.

Category	Definition	Indicator
Pollutant	Pollutant refers to the environmental contaminants produced by transport services, including emissions (such as CO2 and other greenhouse gases) and noise generated during operation.	emission noise
Energy	Energy encompasses the total energy consumed and conserved in transport operations. Indicators such as energy consumption and energy saving assess the system’s environmental footprint and efforts to reduce resource usage.	energy consumption energy saving
Degradation	Degradation measures the natural environmental damage caused by transport infrastructure and activities, including the reduction in land resources, such as trees, fertile land, and water, caused by transport infrastructure development and service operations.	land degradation

.

The specific descriptions of each indicator are as follows.

For Pollutant:

• Emission measures the output of various pollutants generated by transit operations, including specific emissions like CO₂, NOx, PM10, and greenhouse gases. Indicators in this category assess emissions per kilometer traveled, fuel consumption rates, and air quality impacts (measured in parts per million-vehicles) [6,25,30,36,38,44,55,59,60,75,76,82,83].
• Noise quantifies the level of noise pollution generated by service vehicles operating on transit routes, measured in decibels per minute per vehicle [44].

For Energy:

• Energy consumption measures the total amount of energy used in transit operations, reflecting the system’s energy demands and overall efficiency [6,11,32,55,57,68].
• Energy saving captures the extent of energy conserved through various operational efficiencies and improvements [37,76,82].

For Degradation:

• Land degradation measures the dollar value of the reduction in land resources, such as trees, fertile land, and water, caused by transport infrastructure development and service operations [44].

While reducing pollution and energy consumption is crucial for mitigating the environmental impact of public transport systems, it is equally important to avoid efforts that solely focus on minimization of public transport without considering broader implications. Simply reducing transport services might achieve similar outcomes in terms of emissions or energy savings but could compromise service quality. Furthermore, expanding public transport services might initially increase pollution or energy use within the system, but it can lead to a net reduction in overall transportation-related emissions by shifting travelers away from more polluting modes. Therefore, a more holistic perspective that evaluates the environmental impact of the entire transportation network is essential, ensuring that reductions in emissions and energy use are balanced with maintaining the quality of public transport services.

3.4.2. Economic Impact

Economic impact indicators focus on how the system contributes to regional economic development. The indicators are organized into categories, with each category containing specific definitions. The classification of indicators for economic impact is shown in

Table 6. Classification of indicators for economic impact.

Category	Definition	Indicator
Economic Growth	Economic growth refers to the contribution of transport services to the broader economy, including indicators such as employment, tax revenue, and GDP. These indicators evaluate the system’s role in stimulating economic development and supporting regional prosperity.	employment tax revenue GDP growth
Property Impact	Property impact measures the influence of transport services on property values within the service area. Indicators such as house price and land price reflect how the presence of transit services can affect real estate markets and attract investment.	house price land price

.

The specific descriptions of each indicator are as follows.

For Economic Growth:

• Employment measures the total job availability within the transport network’s influence areas, including employment within the catchment area and corridor. It also tracks net job growth by calculating the difference between jobs created and jobs lost [38,60,80].
• Tax revenue measures the increase in tax contributions generated from transport operations and commercial business, based on financial statements from transit operators and tax revenue from commercial activities [38,80].
• GDP growth measures the percentage growth in regional economic output that can be attributed to the transport project. It compares the increase in regional GDP after the project’s implementation to a scenario where the project was not constructed [37].

For Property Impact:

• House price measures how public transport affects both residential and commercial property values in the influenced area. Indicators include changes in average house prices and commercial property prices, often assessed through hedonic pricing models considering factors like location, size, accessibility, and transport proximity [31,37,38].
• Land price reflects the impact of public transport systems on land values, typically evaluated using hedonic pricing models based on panel surveys [80].

3.4.3. Societal Impact

Societal impact indicators focus on how the system promotes social inclusion and benefits communities. The indicators are organized into categories, with each category containing specific definitions. The classification of indicators for societal impact is shown in

Table 7. Classification of indicators for societal impact.

Category	Definition	Indicator
Accessibility	Accessibility evaluates the ease with which users can reach the transport service (i.e., access to transit) and use the service to reach their destinations (i.e., access by transit).	access to transit access by transit
Affordability	Affordability evaluates the economic feasibility of transport services by measuring the cost burden of mobility relative to users’ income and the availability of affordable living conditions within the transit influence area.	transport cost affordability housing cost affordability
Inclusivity	Inclusivity evaluates how well the transport system serves specific underserved or vulnerable groups, such as low-income residents, seniors, or individuals with limited mobility.	coverage inclusivity physical inclusivity
Equity	Equity measures fairness in distributing transport services and resources across population groups and geographic areas. It includes fair access to transit services, equitable cost distribution among users, and balanced allocation of transport investments and subsidies	access equity fare equity resource allocation
Community Benefit	Community benefit measures the positive impacts of transportation systems in two areas: reducing traffic congestion and promoting physical health through active commuting.	congestion relief health benefit

.

The specific descriptions of each indicator are as follows.

For Accessibility:

• Access to transit measures the spatial and temporal reach to transport services within a region, considering factors such as travel distance, walking time, and travel resistance to the nearest transit stops [23,33,35,84,85,86]. It is also used as an indicator to estimate social exclusion based on the maximum distance that someone has to walk to reach public transport [87].
• Access by transit refers to the ability of individuals to reach their destinations using public transport. It is often measured using gravity-based models or cumulative opportunity approaches, incorporating factors such as spatial distribution of destinations (e.g., jobs, education, healthcare) and distance decay effects [28,84,86,88,89,90,91].

For Affordability:

• Transport cost affordability measures the proportion of household or personal income spent on transportation, assessing whether transit services are economically sustainable for users [38,92].
• Housing cost affordability evaluates the presence of affordable housing options within the transit corridor influence area [38].

For Inclusivity:

• Coverage inclusivity focuses on the inclusiveness of public transport service coverage, ensuring that disadvantaged or marginalized groups have equitable access to essential transit options and services within their regions [38,53].
• Physical inclusivity focuses on the design of transport services, ensuring they accommodate individuals with physical disabilities [30].

For Equity:

• Access equity measures the fairness in the distribution of accessibility to transit services among different population groups and geographic areas. It evaluates disparities using methods such as the Lorenz curve, Gini index, and normalized accessibility indices, ensuring accessibility aligns with socioeconomic needs and critical life opportunities [28,35,38,84,85,89].
• Fare equity evaluates the fairness in cost distribution among passenger groups, ensuring that users pay proportionally to the services they consume [64,93].
• Resource allocation evaluates the distribution of transport investments and subsidies, particularly focusing on the proportion allocated to remote areas, to ensure equitable support for geographically disadvantaged regions [55].

For Community Benefit:

• Congestion relief measures the decrease in traffic congestion through public transportation, indicated by indicators like avoided vehicle-kilometers and reduced traffic during peak hours [52,74].
• Health benefit evaluates the contribution of public transport to individual physical health by measuring the energy expended through physically active transport modes, such as walking or cycling [79]. Although health is also influenced by factors like pollutant emissions, traffic accidents, and walking time, these indicators are categorized separately under environmental, safety, and time indicators, respectively, and are not included in this category.

The detailed categorization and descriptions of the organized indicators are provided in the Supplementary Materials for further reference.

3.5. Comparative Classification and Outcome-Oriented Synthesis

Section 3.3 and Section 3.4 present a detailed catalog and definitions of indicators for both direct and indirect evaluation. To enhance readability and provide a clear overview, these descriptions are synthesized into a concise classification in

Table 8. Concise classification of public transport evaluation indicators.

Main Category	Subcategory	Indicators
Supply–Planning	Network Design	stop density; network density; connectivity; network length; number of routes; coverage; route directness; number of stops; centrality; route overlap; number of intersections
	Resource	fleet size; fleet features; fleet age; number of employees
	Finances	operational cost; capital cost; total cost; subsidy
	Service Provision	frequency(scheduled); operating time (scheduled)
Supply–Operating	Service Provision	operating length; frequency (delivered); operating time (delivered); active fleet size; passenger capacity
	Time	headway; trip time; running time; dwell time; layover time
	Reliability	punctuality; regularity; delay; disruption
	Speed	commercial speed; running speed; peak-hour speed; acceleration; deceleration
	Finances	operational cost; incident cost
Demand–Planning	Socioeconomic Information	land use; population
	Potential Travel Needs	demand coverage; user-group characteristics
Demand–Operating	Service Utilization	ridership; load factor; number of trips; passenger-kilometers; mode share; travel distance; active hours; empty distance; complaints; trip completion rate; mode choice probability
	Time	wait time; travel time; in-vehicle time; out-of-vehicle time; time saving
	Finances	revenue; fare
	Transfer	transfer time; number of transfers; transfer ridership; transfer rate
	Safety	number of accidents; accident frequency; number of casualties; number of crimes
Environmental Impact	Pollutant	emission; noise
	Energy	energy consumption; energy saving
	Degradation	land degradation
Economic Impact	Economic Growth	employment; tax revenue; GDP growth
	Property Impact	house price; land price
Social Impact	Accessibility	access to transit; access by transit
	Affordability	transport cost affordability; housing cost affordability
	Inclusivity	coverage inclusivity; physical inclusivity
	Equity	access equity; fare equity; resource allocation
	Community Benefit	congestion relief; health benefit

.

While Table 8 provides a systematic overview of how indicators can be organized by supply, demand, and indirect impacts, this perspective also creates a foundation for considering broader policy objectives. Recent evaluation frameworks increasingly emphasize sustainability goals such as environmental protection, social inclusion, and economic development [94]. To capture these higher-level policy concerns, it is useful to highlight which subsets of indicators relate to sustainability outcomes. Building on the classification, we therefore group representative indicators into three clusters (environmental, social, and economic) in

Table 9. Key indicator clusters relevant to sustainability outcomes.

Cluster	Representative Indicators	Notes on Relevance
Environmental	emissions; energy consumption; noise; operating time; vehicle-kilometers; fleet size	Environmental sustainability reflects how transport operations influence ecological quality and resource use. Emissions and noise directly capture the environmental externalities of transport activities, while energy consumption indicates the system’s demand on natural resources. Operational factors such as vehicle-kilometers, fleet size, and operating time link service supply to environmental burden since more vehicles and longer operations typically increase pollution and resource use. At the same time, efficient fleet deployment and optimized service hours can mitigate negative impacts, balancing service provision with ecological preservation.
Economic	revenue; operational cost; employment; property impacts; frequency; load factor	Economic sustainability reflects whether transport systems create financial stability for operators and stimulate broader regional development. Revenue and operational cost capture financial viability, showing if services can be maintained without unsustainable subsidies. Employment and property impacts link transport directly to regional growth by generating jobs and enhancing land and housing values. Service frequency and load factors connect operational decisions with productivity, as they influence how efficiently resources are used and how effectively the system supports economic activity.
Social	access to transit; travel time; reliability (punctuality, delay); affordability (fare); inclusivity (coverage, physical)	Social sustainability reflects how transport systems support equity, inclusion, and quality of life for diverse population groups. Indicators such as accessibility and inclusivity directly measure whether vulnerable or underserved populations can reach essential destinations and participate in society. Reliability and travel time influence people’s ability to depend on public transport for daily activities, shaping confidence and trust in the system. Fare levels capture affordability, ensuring that mobility does not create excessive financial burden and allowing equitable participation in social opportunities.

.

4. Towards Evaluation of Advanced Mobility Systems

4.1. Relevance of Existing Indicators to Advanced Mobility Systems

The previous chapter provided a comprehensive overview of existing public transportation evaluation indicators. However, the emergence of advanced mobility systems introduces unique features and challenges that may require tailored evaluation indicators. In this section, we analyze six advanced mobility services: Demand Responsive Transport (DRT) [5,6], shared mobility [7], micromobility [8], autonomous mobility [9], Personal Rapid Transit (PRT) [10], and Mobility as a Service (MaaS) [11]. For each service, we discuss its defining characteristics, propose evaluation considerations from the perspectives of users, operators, and administrators (e.g., city governments), and examine the relationship between existing indicators and those needed to fully capture the capabilities and impacts of advanced mobility systems. In doing so, we classify indicators into three categories: “Adapted” indicators, which can be applied with minor contextual adjustments; “Transformed” indicators, which require redefinition to reflect flexible and dynamic operations; and “New” indicators, which capture unique features not addressed by conventional measures. It is important to note that these services are not entirely distinct, rather, the discussion highlights their most representative characteristics while acknowledging overlapping features and shared goals.

4.1.1. Demand Responsive Transport (DRT)

Demand Responsive Transport (DRT) is a flexible transportation model that dynamically adjusts routes and schedules based on real-time user demand. It is particularly suitable for low-demand areas, non-peak hours, or for serving user groups requiring customized transportation solutions [6]. Unlike traditional fixed-route systems, DRT eliminates pre-defined routes and timetables, offering a personalized and adaptive service. However, the level of flexibility may vary across different types of DRT services, ranging from fully responsive systems to semi-dynamic ones, depending on operational design and user needs.

From a user perspective, the evaluation focus is whether DRT meets their travel needs by providing access to destinations at preferred times in traditional public transportation systems, with indicators such as delay in Section 3.3.2 evaluating service reliability by comparing actual and scheduled arrival times. However, in DRT, the user-expected arrival time depends on individual needs rather than fixed schedules. This creates additional considerations, such as waiting time, which arises from the absence of fixed timetables, making users uncertain about when a vehicle will arrive. Waiting time can be evaluated using the existing “wait time” indicator in Section 3.3.4. Additionally, the flexible and shared nature of DRT routes can lead to an increase in travel time when vehicles make detours to serve other passengers before reaching an individual’s destination. This can build on the “time saving” indicator in Section 3.3.4 but is adjusted to account for added travel time rather than reductions. To further capture the user experience, we consider the difference between actual and user-expected arrival time as a new indicator to capture the gap between the system’s performance and user expectations alongside the probability of unmet user requests, which is considered to measure how often the system fails to meet user needs, which traditional dial-a-ride systems cannot measure.

From an operator perspective, the evaluation focuses on demand fulfillment, operational efficiency, and economic sustainability. The proportion of user-demand fulfilled builds on existing indicators such as demand coverage in Section 3.3.3, refining it to assess individual service requests rather than aggregate passenger counts. Similarly, the ratio of empty-drive distance to total is an existing indicator from Section 3.3.4, suitable for evaluating operational efficiency in both fixed and flexible systems. To further capture operational efficiency specific to DRT, we introduce the shared-ride rate as a new indicator to assess the system’s ability to optimize vehicle usage by matching multiple users with overlapping travel needs. Additionally, the fleet size indicator from Section 3.3.1 is relevant, as DRT requires an optimal number of vehicles. Too few vehicles may fail to meet demand, while too many could lead to resource inefficiency. Economic sustainability continues to be assessed through the ratio of operating cost to revenue, also detailed in Section 3.3.4.

From an administrator perspective, ensuring equitable service distribution and flexibility across regions is critical. We consider using an indicator of the spatial distribution of service utilization to evaluate whether DRT vehicles effectively serve all areas or are concentrated on a few fixed routes. This ensures that DRT complements traditional public transportation by reaching underserved areas where conventional transit is less effective. This builds on concepts from access equity in Section 3.4.3 Societal Impact but shifts the focus to the balance of route utilization across regions.

Table 10. Summary of indicators for DRT.

Perspective	Evaluation Focus	Indicators	Relationship	Explanation
user	Ability to meet travel needs	Waiting time	Adapted	In DRT, the absence of fixed timetables makes passengers uncertain about vehicle arrival; the conventional “waiting time” indicator remains valid to evaluate this aspect.
		Increase in travel time	Transformed	Detours to pick up additional passengers may increase individual travel times; this builds on the conventional “time saving” indicator but is redefined in the opposite direction.
		Difference between actual and user-expected arrive time	Transformed	Conventional “delay” measures deviations from scheduled times; in DRT, expected arrival is user-defined, requiring a redefinition as the gap between actual and user-expected arrival times.
		Probability of unmet user requests	New	Captures the frequency of service failures unique to DRT, which cannot be measured in fixed-route systems.
operator	Demand fulfillment	Proportion of user-demand fulfilled.	Transformed	Extends the traditional “demand coverage” concept by focusing on the fulfillment of individual service requests rather than aggregate passenger counts.
	Operational efficiency	Ratio of empty-drive distance to total	Adapted	A well-established indicator of operational efficiency, directly applicable to both fixed-route and flexible systems.
		Shared-ride rate	New	Reflects the efficiency of dynamic ride-matching in DRT by measuring the extent of shared rides, absent in fixed-route evaluations.
		Fleet size	Adapted	A conventional indicator of resource allocation; still critical in DRT, where too few vehicles may fail to meet demand and too many leads to inefficiency.
	Economic sustainability	Ratio of operating cost to revenue	Adapted	A standard financial performance indicator that remains valid for evaluating DRT.
administrator	Service flexibility in usage	Spatial distribution of service utilization	Transformed	Builds on the concept of access equity, but in DRT, it shifts the focus to balancing service deployment across regions, ensuring coverage beyond fixed corridors.

summarizes the evaluation focuses, corresponding indicators for DRT, and their relationships with existing indicators.

4.1.2. Shared Mobility

(1)

Ride share

Ride share encompasses diverse implementations. In Japan, there are two distinctive models: Japan-style ride-share systems and public ride-share systems [95]. Japan-style ride-share systems involve utilizing private vehicles or part-time drivers to address taxi shortages in specific regions and times, under the regulation of local taxi companies. Public ride-share systems, on the other hand, are organized by municipalities or non-profits to fill transportation gaps in underserved areas, prioritizing community welfare over profit, making them somewhat similar to DRT. Beyond these, a more popular model globally is Commercial Ride Share, which is operated by profit-driven Transportation Network Companies (TNCs) like Uber and Lyft, which leverage dynamic pricing and real-time route matching to maximize revenue while meeting passenger demand. In this research, we focus on the generalized concept of individuals using their private vehicles during their spare time to provide mobility services, regardless of distinctions such as profitability, organizing entities, or geographic coverage.

From a user perspective, ride share takes place in a relatively private space, which can influence perceptions of safety. Studies have shown that women may feel less safe and comfortable when sharing rides with unknown male passengers or during nighttime trips [96]. One relevant indicator is the presence of in-vehicle safety measures, such as surveillance cameras, emergency communication systems, or driver screening protocols. This indicator aligns with fleet features from Section 3.3.2, which evaluates vehicle characteristics. Additionally, travel efficiency, encompassing factors such as waiting time and increased travel time due to shared routes, has already been discussed in the DRT section, but it remains relevant here as it similarly impacts user experience and can be assessed using existing indicators with modifications for ride-share scenarios.

From an operator perspective, ride share focuses on improving ride-matching efficiency and optimizing vehicle utilization. The ride-matching success rate indicator measures the system’s ability to consolidate passenger trips effectively, which is absent in traditional public transit systems, as they lack dynamic matching requirements. The load factor, an existing indicator from Section 3.3.4, reflects vehicle capacity utilization and directly impacts revenue since operational costs remain relatively stable for one shared trip regardless of the number of passengers.

From an administrator perspective, ride share contributes to alleviating congestion by reducing single-occupancy vehicle trips and reducing vehicle miles traveled, which reflect ride share’s contributions to societal goals. These contributions can be evaluated using existing indicators, such as congestion relief and modal share from Section 3.4.3 Societal Impact, with adjustments to reflect ride share’s specific impacts. It is important to note that ride share may also lead to unintended side effects, such as a decrease in the demand for traditional public transport services like taxis [97], potentially disrupting the mobility system balance. This can be evaluated using the taxi demand decrease indicator, building on the mode share indicator in Section 3.3.4 but focusing on the reduction aspect.

Table 11. Summary of indicators for shared rides.

Perspective	Evaluation Focus	Indicators	Relationship	Explanation
user	Safety in private spaces	Presence of in-vehicle safety measures	Transformed	Aligns with the conventional “fleet features” idea but reorients it to safety attributes given the need for user assurance and safety in ride-share environments.
	Travel efficiency	Waiting time	Adapted	Still needed to assess user experience; waiting time remains relevant and can be assessed with existing indicators.
		Increase in travel time	Transformed	Reflects detours due to shared routes; conceptually builds on “time saving” but is redefined in the opposite direction, as discussed in DRT.
operator	Ride-matching efficiency	ride-matching success rates	New	Captures dynamic matching performance absent in traditional fixed-route transit.
	Vehicle utilization	Load factor	Adapted	Standard capacity-utilization indicator that remains applicable.
administrator	Congestion mitigation	Reduction in single-occupancy trips	Adapted	Uses existing societal indicators (e.g., congestion relief, modal share) to reflect how ride share reduces solo driving.
		Reduction in vehicle miles traveled	Adapted	Same existing societal indicators apply, with scope adjusted to ride-share impacts.
	Side effect	Taxi demand decrease	Transformed	Builds on “mode share” but focuses on the reduction specific to taxis as an unintended system impact.

summarizes the evaluation focuses, corresponding indicators for ride share, and their relationships with existing indicators.

(2)

Shared vehicles

Shared vehicles provide users with access to various vehicle types, such as cars, bikes, or scooters, allowing them to drive or ride exclusively as needed. While both shared vehicles and ride share aim to maximize resource utilization, their operational modes differ. Ride share involves multiple passengers sharing a vehicle simultaneously, while shared vehicles are used by different users at different times, with vehicles flowing between users. This flexibility reduces the necessity of vehicle ownership.

From a user perspective, the evaluation focuses on the availability of vehicles when needed. For example, private car owners would only consider offering their car as a shared vehicle if they could reliably access it whenever required. Research has also shown that the availability of shared bicycles significantly influences users’ brand preferences [98]. Therefore, ensuring sufficient vehicle availability at convenient times and locations is critical. We proposed a new indicator, vehicle availability at time of request, that measures the percentage of user requests relating to vehicle availability and accessibility within a specified time frame (e.g., 5 min) from the time of request, which traditional public transportation systems do not cover. Additionally, vehicle features such as cleanliness and the availability of navigation systems influence users’ choice of vehicles, which can be evaluated using the fleet feature indicator in Section 3.3.1.

From an operator perspective, the focus is on maximizing utilization and minimizing idle time, as continuous vehicle usage is key to operational efficiency. Idle vehicle time, similar to an existing indicator of empty-drive distance in Section 3.3.4 Demand–Operating, reflects how often vehicles remain unused and directly impacts overall productivity. Additionally, since shared vehicles are user-driven and often experience extended driving periods and diverse usage patterns, indicators built on operational cost in Section 3.3.2 but requiring extension to include relocation expenses become critical for ensuring vehicles remain operational.

From an administrator perspective, shared vehicles contribute to societal benefits by reducing the overall demand for vehicle ownership. If a shared vehicle is primarily used by the same individual, it undermines its purpose of reducing vehicle ownership. Unlike traditional public transportation, which evaluates ridership without considering whether the same individual repeatedly uses the service, shared vehicles aim to encourage diverse usage. Therefore, an indicator like the average number of unique users per vehicle per day is critical to help evaluate whether shared vehicles are being used equitably. Similarly, depot equity plays a significant role in ensuring fair accessibility to shared vehicle depots across different regions, particularly underserved areas. The indicator “access equity” in Section 3.4.3 can be adapted to evaluate whether depot locations equitably meet user needs. Additionally, since shared vehicle users act as their own drivers, unlike professional drivers in public transport, user safety education becomes essential, alongside tracking existing indicator accident frequency, to ensure safe operations.

Table 12. Summary of indicators for shared vehicles.

Perspective	Evaluation Focus	Indicators	Relationship	Explanation
user	Availability of vehicles when needed	Instantaneous availability rate	New	Measures the share of requests where a vehicle is available within a set time window at the time of request—an availability-at-request indicator not covered by conventional public transport.
	Vehicle features	Fleet feature	Adapted	Vehicle characteristics continue to shape user choice and can be evaluated with the existing fleet features indicator.
operator	Efficiency of vehicle utilization	Vehicle empty time	Transformed	Mirrors the existing “empty-drive distance” concept to capture periods when vehicles sit unused.
	Maintenance	Maintenance costs	Transformed	Builds on operational cost but requires extension to account for relocation expenses typical of user-driven, distributed assets.
administrator	Equity of usage	Average number of unique users per vehicle per day	New	Unlike traditional ridership totals, this indicator checks whether vehicles serve diverse users (not repeatedly the same individual), aligning with ownership-reduction goals.
	Equity for depot	Equity in depot placement	Adapted	Adapts “access equity” to assess whether depot locations equitably meet user needs.
	User safety	Level of user safety education	New	Required because users act as their own drivers, alongside tracking existing indicator accident frequency, to ensure safe operations.
		Accident frequency	Adapted	An established safety indicator that remains applicable to monitor incident rates under shared-vehicle operations.

summarizes the evaluation focuses, corresponding indicators for shared vehicles, and their relationships with existing indicators.

4.1.3. Micromobility

Micromobility refers to lightweight, low-speed vehicles such as bicycles, electric scooters, and velomobiles, designed for short-distance travel [8]. These vehicles are typically shared or rented for brief periods and cater to first-mile and last-mile connectivity or short urban trips. Unlike shared mobility, micromobility is primarily focused on personal, short-distance travel, offering flexibility and convenience while reducing reliance on private vehicles.

From a user perspective, micromobility evaluates availability and physical accessibility. While similar to shared vehicles in ensuring sufficient vehicles are present when needed, micromobility emphasizes ease of locating and accessing vehicles within a short walking distance. Key indicators include the density of vehicles in service areas and the average time to locate and activate a vehicle. The density of vehicles builds on the concept of fleet size from Section 3.3.1, focusing on specific smaller service areas rather than overall availability. Similarly, the average time to locate and activate a vehicle aligns with the existing indicator walk time to transit stops from Section 3.3.4 but reflects the dispersed nature of vehicles rather than fixed stations.

From an operator perspective, micromobility systems prioritize fleet management efficiency, particularly addressing the issue of uneven demand. As micromobility is often used as a last-mile solution, vehicles are frequently ridden from high-demand areas (e.g., transit hubs) to low-demand areas (e.g., residential neighborhoods). To avoid vehicles being abandoned in low-demand areas, depot planning efficiency becomes a key indicator. We consider a new indicator, “vehicle turnover rate at depots”—the average number of vehicles redistributed per day per depot—to measure how efficiently vehicles are returned to service from low-demand areas. Even with effective depot planning to ensure efficient redistribution of vehicles during the planning phase, vehicle accumulation in low-demand areas may still require redistribution to high-demand areas during the operational phase, making rebalancing costs an important operational consideration.

From an administrator perspective, micromobility supports first- and last-mile connectivity, encouraging public transportation use. In low-demand areas, maintaining basic services may not be profitable for operators, but governments may prioritize service equity in low-demand areas to ensure access, aligning with the existing indicator access equity in Section 3.4.3. Additionally, the change in public transportation usage tracks how micromobility affects transit ridership, similar to the existing indicator modal share in Section 3.4.3.

Table 13. Summary of indicators for micromobility.

Perspective	Evaluation Focus	Indicators	Relationship	Explanation
user	Availability and accessibility	Density of vehicles in areas	Transformed	Builds on the conventional “fleet size” idea but applied to smaller service areas to ensure vehicles are sufficiently available within short walking distances.
		Average time to locate and activate a vehicle	Transformed	Aligns with the existing “walk time to transit stops”, redefined for the dispersed, non-station nature of micromobility vehicles.
operator	Fleet management efficiency	The average number of vehicles redistributed per day per depot	New	Introduced to capture depot planning effectiveness when demand is uneven and vehicles accumulate in low-demand areas.
		Rebalancing cost	Adapted	An established operational consideration that remains relevant as vehicles are moved from low- to high-demand areas during operations.
administrator	Equity in service provision	Service equity in low-demand areas	Adapted	Adapts “access equity” to emphasize maintaining basic access where operations may be less profitable.
	Public transportation integration	Change in public transportation usage	Adapted	Uses the existing “modal share” perspective to track how micromobility affects transit ridership.

summarizes the evaluation focuses, corresponding indicators for micromobility, and their relationships with existing indicators.

4.1.4. Autonomous Mobility

Autonomous mobility refers to transportation systems operated by self-driving vehicles. These systems aim to improve safety and operational efficiency while reducing the reliance on human drivers [9]. Autonomous mobility is suited for both fixed-route services and dynamic, on-demand operations, offering scalability for various use cases.

From a user perspective, the primary focus is on safety and comfort. For safety, users are concerned with how effectively the system can detect and respond to sudden incidents. We considered a new indicator to measure the response time to unexpected events, defined as the time taken from detecting a hazard to executing an appropriate action, such as braking, steering, or issuing a warning. This is a new indicator not covered by traditional public transportation indicators. For comfort, users prioritize the smoothness of vehicle operation, with a focus on acceleration and deceleration stability. This indicator evaluates whether the system maintains a comfortable ride experience and is also unique to autonomous mobility.

From an operator perspective, the focus lies on user acceptance and operational reliability. User acceptance is critical for system adoption, and the user acceptance survey score serves as a key indicator to assess public attitudes, aligning with similar indicators used in traditional public transportation systems. For operational reliability, the indicator system uptime percentage evaluates the proportion of time the autonomous system is functioning as expected. This is similar to existing indicators like empty distance in Section 3.3.2 Supply–Operating, which measure specific operational states, but it is specifically designed to ensure disruptions. Additionally, maintenance costs become a crucial consideration, as autonomous vehicles often require regular updates and specialized repairs to maintain system functionality. This builds on operational cost in Section 3.3.2 but focuses on the unique needs of autonomous operations.

From an administrator perspective, autonomous mobility offers the potential to enhance the safety of the overall transportation network by reducing human error. The reduction in overall traffic accident rates indicator measures the societal benefits of integrating autonomous vehicles into public transportation systems. This aligns with existing public-safety indicators, focusing on broader improvements to road safety.

Table 14. Summary of indicators for autonomous mobility.

Perspective	Evaluation Focus	Indicators	Relationship	Explanation
user	Safety	Response time to unexpected events	New	Measures the time from detecting a hazard to executing an appropriate action; this aspect is not covered by traditional public transport indicators.
	Comfort	Smoothness of vehicle operation	New	Focuses on acceleration and deceleration stability to ensure a comfortable ride; treated as a unique requirement of autonomous mobility.
operator	User acceptance	User acceptance score	Adapted	Aligns with similar acceptance or attitude measures used in traditional public transport; remains critical for adoption.
	System reliability	System uptime percentage	Transformed	Evaluates the proportion of time the autonomous system functions as expected; similar to the existing operational-state indicator “empty distance”.
	Maintenance cost	System maintenance expenses	Transformed	Builds on operational cost but extended to cover autonomous operation needs (regular updates and specialized repairs).
administrator	Reduction in overall traffic accident rates	Percentage reduction in accidents	Adapted	Uses existing public-safety indicators to capture societal benefits from reduced human error with autonomous mobility integration.

summarizes the evaluation focuses, corresponding indicators for autonomous mobility, and their relationships with existing indicators.

4.1.5. Personal Rapid Transit (PRT)

Personal Rapid Transit (PRT) refers to fully automated, small-scale transit systems operating on dedicated guideways [10]. PRT systems are designed to provide on-demand, point-to-point transportation, typically serving urban areas, airports, or campuses. Unlike traditional fixed-route systems, PRT eliminates the need for schedules or transfers, offering taxi-like experience in a controlled environment.

From a user perspective, the evaluation focuses on service reliability. Since short waiting times are a core competitive advantage of PRT, users prioritize how quickly vehicles arrive after a request is made. The average time from request to vehicle arrival measures this responsiveness, reflecting whether the system meets user expectations for real-time service. While this is conceptually similar to the wait time indicator in Section 3.3.4, it is a new indicator tailored to PRT’s dynamic and on-demand operations. Additionally, the connectivity of access and egress are critical factors influencing the overall user experience. Access refers to the convenience and time required for users to reach PRT stations from their starting points, while egress pertains to the ease of reaching their final destinations from PRT stations. These aspects can be evaluated using the access to transit indicator in Section 3.4.3 but should also include “transit access to destination” to comprehensively capture both ends of the journey.

From an operator perspective, the focus is on ensuring system stability, optimizing vehicle utilization, and the costs associated with dedicated infrastructure. The system stability indicator, which assesses whether operations are consistently running without disruptions, is essential for maintaining continuous service. This indicator is similar to the accident frequency measure in Section 3.3.4 but includes minor interruptions and delays. Additionally, PRT’s small vehicle capacity necessitates efficient scheduling and dispatch to avoid excessive waiting times for users. The load factor evaluates how effectively vehicle capacity is utilized, and this is an existing indicator commonly used in public transit systems. Furthermore, because PRT relies on dedicated guideways, the construction and maintenance costs of these guideways are critical. The capital cost indicator, which tracks expenses for infrastructure, is another relevant measure.

From an administrator perspective, PRT serves as an essential supplement to the urban transportation system, offering a vital connection for microcirculation within cities. A key focus for administrators is whether PRT improves the overall efficiency of the transportation network. The average trip time across the entire network measures how PRT integration affects the urban system’s performance, ensuring it aligns with broader goals of reducing travel times. While this is similar to the travel time indicator in Section 3.3.4, it specifically evaluates the impact of PRT on the entire transportation network.

Table 15. Summary of indicators for PRT.

Perspective	Evaluation Focus	Indicators	Relationship	Explanation
user	Service reliability	Average time from request to vehicle arrival	New	Tailored to PRT’s on-demand operations to capture real-time responsiveness; conceptually related to “wait time” but defined as the request-to-arrival interval.
	Connectivity	Access and egress	Transformed	Builds on “access to transit” but extended to cover both station access and egress to final destinations.
operator	System stability	System disruption rate	Transformed	Similar to “accident frequency” but broadened to include minor interruptions and delays that affect continuous operation.
	Vehicle utilization	Load factor	Adapted	Established capacity-utilization indicator that remains applicable; especially relevant given PRT’s small vehicle capacity and need for efficient dispatch.
	Infrastructure costs	Capital cost	Adapted	Existing indicator tracking construction and maintenance expenses for dedicated guideways.
administrator	Network efficiency improvement	Average trip time across the entire network	Transformed	Related to “travel time” but redefined at the network scale to assess PRT’s system-wide impact on average trip time.

summarizes the evaluation focuses, corresponding indicators for PRT, and their relationships with existing indicators.

4.1.6. Mobility as a Service (MaaS)

Mobility as a Service (MaaS) integrates various transportation modes into a single digital platform that allows users to plan, book, and pay for trips seamlessly [99]. By emphasizing convenience and connectivity, MaaS aims to reduce reliance on private vehicles and promote sustainable transportation through multimodal integration.

From a user perspective, the evaluation focuses on accessibility and ease of use. Users expect MaaS platforms to provide seamless trip planning, real-time updates, and transparent pricing across multiple transportation modes. Indicators such as total travel time across modes, waiting time, and ease of payment are critical for assessing platform usability. The total travel time and waiting time aligns with the travel time indicator and wait time indicator from Section 3.3.4, while ease of payment reflects elements of fleet feature in Section 3.3.1 Supply–Planning but extends to include digital services like app usability.

From an operator perspective, MaaS emphasizes system integration and service optimization. Integration is measured by the number of transportation modes successfully incorporated into the platform, reflecting the platform’s ability to connect diverse services, a new indicator unique to MaaS. Additionally, multimodal transfer efficiency, which evaluates the smoothness of transitions between modes, is similar to transfer time in Section 3.3.4 but applies to multiple transportation systems.

From an administrator perspective, MaaS aims to promote sustainable transportation and equity. The public transport usage growth rate reflects efforts to encourage a modal shift from private vehicles and aligns with modal share in Section 3.4.3. To ensure inclusivity, the user coverage rate (e.g., by age, income, location) and the presence of accessibility-friendly design are critical. These indicators relate to access equity and fleet features in Section 3.4.3 and Section 3.3.1, respectively, but focus on digital inclusivity.

Table 16. Summary of indicators for MaaS.

Perspective	Evaluation Focus	Indicators	Relationship	Explanation
user	Ease of use across multiple modes	Total travel time across modes	Adapted	Aligns with the conventional “travel time” indicator, applied to end-to-end multimodal itineraries on a MaaS platform.
		Waiting time	Adapted	The standard “waiting time” measure remains valid for assessing usability across chained modes.
		Ease of payment	Transformed	Related to “fleet features” but extended to digital services; specific to MaaS platforms.
operator	System integration	Number of transportation modes successfully integrated	New	Introduced to capture the platform’s ability to connect diverse services—an integration metric unique to MaaS.
	Service optimization	Multimodal transfer efficiency	Transformed	Similar to “transfer time” but defined across multiple systems to evaluate cross-mode handoffs within MaaS.
administrator	Modal shift from private vehicles	Public transport usage growth rate	Adapted	Tracks induced shift toward public transport and aligns with modal share-based evaluation.
	Digital inclusivity	User coverage rate (age, income, location)	Adapted	Relates to access equity, focusing on whether different population groups are reached via MaaS.
		Presence of accessibility-friendly design	Transformed	Related to “fleet features” but reoriented to digital inclusivity.

summarizes the evaluation focuses, corresponding indicators for MaaS, and their relationships with existing indicators.

4.2. Selecting Indicators for Simulation in Advanced Mobility Contexts

In the previous section, we discussed the aspects and indicators that should be used to evaluate advanced mobility systems. Traditionally, methods such as experiments or surveys are employed to collect data for evaluation. However, these approaches often require significant resources and time. A more economical and widely adopted alternative is a simulation, which allows for preliminary assessment of advanced mobility systems’ performance under various scenarios. In this section, we examine how simulation uses different indicators to evaluate advanced mobility systems from three perspectives: planning, operating, and indirect impacts. Planning indicators act as inputs, providing the necessary data and configurations to set up simulation models, while operating indicators and indirect impacts serve as outputs, reflecting the performance and broader consequences of the systems being evaluated. It is important to note that autonomous mobility is not included in this discussion. Unlike other modes, autonomous mobility is a vehicle operation technology rather than a distinct service model. Evaluating autonomous mobility requires specialized tools and testing environments for autonomous driving systems, which are beyond the scope of this section.

4.2.1. Planning Indicators for Setting up Simulation Parameters

Planning indicators serve as foundational settings that establish the structural parameters of the simulation environment, defining core aspects such as service areas and demand distribution. These indicators create the baseline conditions for evaluating each advanced mobility mode.

Supply–planning indicators focus on defining the service structure and resource allocation within the simulation:

• Service area: The service area is a critical parameter across all advanced mobility systems, as it establishes each system’s spatial reach and coverage. For MaaS, the service area is not only the areas covered by advanced mobility modes but also incorporates traditional public transport networks, including their routes and schedules (network design information). In simulation environments, the service area is set as the permitted operating geography by specifying polygons or zone identifiers, the allowed subset of transport network links and nodes, and station or dock locations within which vehicles may be dispatched and operate.
• Fleet size: Fleet size indicates the number of vehicles or units available for a given mobility mode, a fundamental input for all advanced mobility systems. By adjusting fleet size, simulations can model various levels of service availability and test scenarios with different resource allocations. For instance, in micromobility systems, density of vehicles in areas serves as a localized extension of fleet size, emphasizing resource placement to maximize accessibility. In simulation environments, fleet size is set by declaring the total number of vehicles per mode. Scenarios then vary this parameter to examine its effects on performance.
• Fleet distribution: For shared vehicle and micromobility, the distribution of fleet resources across service areas is essential. Indicators such as density of vehicles in areas refine fleet distribution by focusing on localized resource availability, enabling simulations to assess service accessibility in specific areas. Additionally, for micromobility, depot planning efficiency ensures vehicles are optimally placed to minimize vehicle idle time and maximize system efficiency. These systems benefit from a distribution plan that allocates resources to areas with the greatest demand rather than focusing solely on fleet size. In simulation environments, fleet distribution is set by specifying initial inventories per zone, station, or depot, capacity limits, and relocation or rebalancing rules.
• Operating time: The operating time specifies the hours during which services are available. For MaaS, the operating time includes traditional public transport services and their scheduled frequencies. Therefore, MaaS requires a broader service provision parameter, encompassing not only the availability of advanced mobility modes but also the integration of traditional modes to ensure seamless transitions between services. In simulation environments, the operating time is set by declaring service hours per mode, first and last trips and scheduled headways for fixed-route services, and dispatch or acceptance windows for on-demand services; these settings govern when agents can request, transfer, and travel.
• Financial indicators: Financial indicators, including fare structure, are essential for modeling both demand and revenue. Adjusting fare levels allows the simulation to evaluate how price variations impact user demand across different income groups and trip purposes. For example, operational cost parameters within the simulation can predict the financial sustainability of different advanced mobility modes. In systems like PRT, capital costs, such as the expenses associated with constructing dedicated guideways, vary significantly depending on the extent of the network and must be accounted for in planning scenarios. In simulation environments, financial indicators are set by defining fare tables (flat, zonal, distance, or time based), discounts and transfer rules, payment caps or passes, and any dynamic pricing; cost parameters are specified for vehicle-hours, vehicle-kilometers, energy and labor, and capital costs so that revenue and cost outputs can be computed consistently.

Demand–planning indicators focus on analyzing potential demand precisely:

• Demands: In traditional transit planning, demand analysis is often based on population data and land-use statistics. However, advanced mobility systems can benefit from demand analysis methods that incorporate real-world data sources such as census data, public transport data, and big data from mobile or online platforms. These data-driven indicators allow for more accurate demand estimates that reflect actual travel behavior rather than relying solely on demographic assumptions. For example, in simulation environments, demand can be set by specifying origin–destination matrices or request streams (by time of day and day of week), sampling trip attributes from empirical distributions (purpose, party size, trip length), and applying scenario multipliers or growth factors. In addition, user preferences for advanced mobility systems also influence demand. These preferences can be captured through stated preference (SP) surveys, which provide direct qualitative insights into user attitudes and willingness to adopt such systems. However, since user preferences are qualitative data, they cannot be directly incorporated into simulations. Section 4.3 further explores the use of alternative indicators to quantitatively approximate subjective factors, providing a more suitable approach for simulations.
• Proportion of user demand fulfilled: Traditionally, this indicator evaluates the ratio of potential demand (e.g., based on population or land-use data) to the actual service provided by the system. In simulation, it is adapted to measure the proportion of specific user requests fulfilled by the system. For example, in DRT, the probability of unmet user requests quantifies the percentage of requests that the system fails to accommodate for individual users, where a request can be recorded as unmet when no feasible vehicle can be assigned within the user’s maximum wait and service constraints (time window, capacity, service area). Similarly, in ride-share systems, ride-matching success rates evaluate the system’s efficiency in consolidating trips, calculated as the proportion of user requests successfully grouped with other passengers, counting a request as matched when the assigned vehicle carries the passenger concurrently with at least one other passenger for any segment of the trip.

4.2.2. Operating Indicators for Evaluating Simulation

Once planning indicators have set up the framework, operating indicators dynamically evaluate system performance as it responds to simulated conditions. Given the diverse array of operating indicators available, selecting the most relevant indicators depends on the specific aspects of performance the simulation intends to evaluate. Below, we outline the operating indicators that are suitable for use within a simulation environment.

Supply–operating indicators focus on the provision of service within the simulation:

• Service provision: This indicator includes both the initial planning phase and ongoing evaluation within the simulation. While the baseline setup defines parameters such as service area, fleet size, and operating hours, dynamic indicators are essential to evaluate system performance during operation. For instance, in DRT, the ratio of empty-drive distance to total distance measures operational efficiency by assessing how much of the vehicle’s travel is unproductive. Similarly, in shared vehicle systems, instantaneous availability rate evaluates the proportion of times a user can access a vehicle immediately when needed, while vehicle empty time measures the duration vehicles remain idle. In simulation environments, empty-drive distance and idle time can be derived from vehicle trajectories and occupancy records, whereas instantaneous availability is assessed at each request event by checking whether a free vehicle is available within the defined radius and time window.
• Time-related indicators: Traditional indicators such as schedule adherence are less applicable for DRT and shared mobility modes, where fixed schedules do not define the operation. Instead, indicators like delay and disruption frequency are more appropriate for capturing the system’s reliability under simulated demand. For DRT, the difference between actual and user-expected arrival time serves as a key indicator to evaluate how closely the service meets individual user expectations. In simulation environments, user-expected arrival time can be read from the trip plan, and actual arrival is taken from drop-off/arrival events. However, for MaaS, schedule adherence remains relevant because it integrates traditional transit modes that operate on fixed schedules alongside flexible modes.
• Speed: Existing speed indicators can be used directly within the simulation to assess performance across various advanced mobility modes. While these indicators may not capture actual vehicle speeds under specific conditions, they can be applied by setting speed parameters based on road type and speed limits. In simulation environments, segment speeds and running/commercial speeds can be obtained from link travel-time outputs and dwell/stop events (including pickup/drop-off time, where relevant), with speed caps set by link attributes or scenario rules.
• Financial indicators: Due to the limitations of simulations in replicating real-world unpredictability, incident-related costs (such as accident costs) are challenging to simulate accurately and are therefore excluded from financial assessment. However, other operational costs can be effectively modeled as part of the simulation process. For example, in shared vehicle systems, maintenance costs can be estimated based on the cumulative distance traveled by vehicles, providing a realistic representation of wear and tear under various usage scenarios. In micromobility systems, rebalancing costs can be calculated based on the number of redistribution operations and the distance covered during these activities. In simulation environments, fares and cost parameters can be specified ex-ante (e.g., per-km, per hour, per-stop, energy), and revenues/costs are aggregated from trip and vehicle logs; maintenance costs are derived from cumulative distance/time per vehicle, while rebalancing costs are tallied from recorded relocation moves and their distances.

Demand–operating indicators focus on evaluating how the system is utilized by users.

• Service utilization: Service utilization is a critical indicator for understanding how advanced mobility systems are used within a simulated environment. For modes like PRT and ride share, load factor, which measures the proportion of occupied seats compared to total vehicle capacity, serves as a key indicator for assessing operational efficiency. This indicator can be calculated from board and alight events or vehicle occupancy timelines over the analysis window in simulation. However, the load factor is not applicable to single-passenger modes such as shared bicycles and e-scooters, where occupancy is inherently limited to one. In such cases, indicators like the average number of unique users per vehicle per day, particularly in shared vehicle systems, provide a more meaningful measure of how effectively vehicles are utilized and shared among users. This indicator can be obtained by counting distinct user identifiers per vehicle within a 24 h period in simulation.
• Time-related indicators: Passenger time indicators, such as travel time and wait time, capture the time spent by passengers during their journey and are applicable within the simulation. For micromobility, the average time to locate and activate a vehicle evaluates how quickly users can begin their trips, reflecting system accessibility. In PRT, indicators such as the average trip time across the entire network and the average time from request to vehicle arrival provide insights into service efficiency and responsiveness. For MaaS, total travel time across modes measures the overall efficiency of integrated systems. In simulation environments, travel time can be recorded as the interval between departure and arrival events, while wait time is the gap between request and pickup. Micromobility locate-and-activate time is measured from vehicle unlock to trip start. For PRT, request-to-arrival time can be the duration between a request submission and the vehicle’s arrival at the boarding point. Network-level averages are computed across all completed trips, and for MaaS, total travel time aggregates all legs and transfers within an itinerary.
• Financial indicators: Financial outcomes, such as total revenue, can be calculated based on simulated usage data. By tracking fares collected from passenger trips across various income levels and trip types, simulations provide insights into the revenue generated by each advanced mobility mode under different scenarios. For instance, the ratio of operating cost to revenue serves as a critical indicator of economic sustainability, assessing whether the revenue generated is sufficient to cover operating expenses. In simulation environments, revenues can be computed by applying fare rules to each completed trip and aggregating results by user or trip attributes. Operating costs can be accumulated from modeled components such as distance, time, energy, and labor. The cost-to-revenue ratio can be then reported for each scenario to evaluate financial viability.
• Transfer-related indicators: Transfer indicators, such as transfer time and the ease of transitioning between modes, are applicable within the simulation. These indicators evaluate how well multimodal systems support seamless connections, reflecting the effectiveness of service integration and user convenience. For MaaS, multimodal transfer efficiency measures the average time and distance required for users to switch modes, providing insights into the smoothness and speed of these transitions. In simulation environments, transfer time can be measured as the interval from alighting one leg to boarding the next, while transfer distance corresponds to the walking path between the two access points. Multimodal transfer efficiency is then summarized as the average of these time and distance values across all simulated itineraries.
• Safety indicators: Safety indicators that rely on incident frequency, such as accident rates, are not feasible in a simulated environment because unexpected events cannot be realistically modeled. For instance, accident frequency and PRT’s system disruption rate, which depend on the occurrence of unplanned incidents, are difficult to evaluate dynamically through simulation. Similarly, shared vehicle indicators like level of user safety education and ride share indicators such as presence of in-vehicle safety measures assess real-world preparedness and vehicle conditions, making them unsuitable for simulation-based evaluation.

4.2.3. Indirect Impact Indicators for Evaluating External Effects

While direct operating and planning indicators capture the immediate performance and setup of advanced mobility systems, indirect indicators provide valuable insights into broader impacts, including environmental, economic, and societal dimensions. These indicators help simulate how advanced mobility systems interact with their surroundings and influence external factors beyond operational efficiency.

Environmental Impact: Environmental indicators in simulation models are typically based on standardized emission rates and energy consumption per unit of distance. By applying these rates, simulations can estimate the environmental footprint of different mobility modes. This method enables a comparative evaluation of various modes’ environmental sustainability, offering insights into pollution levels and energy efficiency. In simulation environments, vehicle trajectories and speeds can be obtained from link-level outputs. Emission and energy factors are applied to each segment, and results are aggregated by trip, by mode, and by spatial units such as grids or zones. The outputs can then be reported as total emissions or energy use for comparison across scenarios.

Economic Impact: Financial impacts within simulations can be partially captured through operational costs and fare revenues, calculated from planning and operating indicators. In simulation environments, fare rules can be applied trip by trip to compute revenues, and cost components can be accumulated from modeled distance, time, labor, and rebalancing. However, advanced economic evaluations require more than these indicators, as they encompass broader regional development, property values, and employment rates. Since simulations primarily model passenger flow and direct costs, they may struggle to capture the full economic impact of advanced mobility systems without additional real-world input. Therefore, evaluating broader economic effects requires integration with external research methods. For instance, hedonic pricing models [100] can estimate the impact of advanced mobility systems on property values, and macroeconomic analyses can assess effects on regional employment and development.

Societal Impact: Societal impact indicators in simulations evaluate whether advanced mobility systems provide accessible, equitable, and inclusive services while contributing to broader transportation goals. Equity is assessed through indicators like the spatial distribution of service utilization in DRT, which examines whether services are evenly distributed across regions, and micromobility’s service equity in low-demand areas, ensuring underserved areas maintain basic service levels. Inclusivity is reflected in MaaS’s user coverage rate, which evaluates the system’s ability to serve diverse populations across different ages, incomes, and locations. Indicators for traffic alleviation, such as ride-share indicators for reduction in single-occupancy trips and reduction in vehicle miles traveled, highlight contributions to reducing congestion. The impact on public transportation usage is measured through micromobility’s change in public transportation usage and MaaS’s public transport usage growth rate, demonstrating how advanced systems encourage shifts from private vehicle use to public transit. In simulation environments, equity can be evaluated by compiling zone-level measures, such as trips served, average waiting time, and vehicle availability, and then analyzing their distribution across space to identify underserved areas. Inclusivity can be examined by using agent attributes such as age, income, and location to compute reach and usage shares for different groups. Congestion-related outcomes can be derived from changes in single-occupancy trips and total vehicle-kilometers compared with a baseline. Changes in public transport usage can be obtained from boarding and alighting events and trip counts by mode, allowing before-and-after comparisons at the corridor or network level.

4.3. Exploring Future Directions for Subjective Aspects in Simulation

While simulations are powerful tools for objectively assessing the operational performance of advanced mobility systems, many existing simulation studies primarily focus on optimizing trips and system efficiency from the operator’s perspective [14]. To ensure a more balanced approach, it is equally important to incorporate user-centered subjective considerations, as factors such as user satisfaction, acceptance, and willingness to adopt these systems significantly impact their real-world success [101]. By including user-focused subjective dimensions alongside operational indicators, simulations can better reflect the complex interplay between operators and users, ultimately supporting the development of mobility systems that align with both technical performance and user expectations.

To incorporate subjective aspects within simulation environments, one effective approach is the use of alternative indicators. Alternative indicators do not directly measure subjective experiences, but they capture factors that are closely related to user perceptions and behaviors, thereby providing an indirect view of user experience. Some examples include the following:

Acceptance and willingness to use: Simulating user acceptance can be challenging, but certain indicators such as frequency of use and repeated trip counts can act as proxies. For example, one study used a roadside Bluetooth sensor network over an extended period to convert pass-by detections into individual trip sequences. From these records, the authors derived repeat-use frequency and regularity by day and time, then grouped users into low, medium, and high repeat clusters. Higher repetition and greater regularity were interpreted as stronger acceptance [102]. In simulation, this idea could translate into logging each agent’s trip histories over a fixed window and summarizing repeat use and temporal regularity, then comparing the distribution of these tiers across scenarios. Another study built an on-demand service experiment in which users submitted origins, destinations, and desired departure times. A request was accepted only when the predicted arrival satisfied a user service-level threshold, such as a maximum wait or a latest acceptable arrival. The authors reported the share of requests that were served and the distributions of waiting time and travel time while varying fleet size, request intensity, and lead time [103]. In a simulation, acceptance can be made measurable by setting the same thresholds, classifying each request as served when the constraints are met and not served otherwise, and then reporting served share and waiting-time percentiles under different scenarios. Similarly, a study on fare-planning work links pricing to choice using a discrete-choice demand model and evaluates alternative fare structures (flat, zonal, distance-based, caps/discounts) by tracking shifts in public-transport choice share, price responsiveness, and user benefit/welfare alongside revenue [104]. In simulation, by implementing the same fare rules and observing how mode or itinerary shares shift by user group and by corridor, larger shares and stronger elasticities could provide quantitative signals of willingness to use under each pricing design.

Satisfaction through convenience proxies: User satisfaction is often closely tied to convenience, which can be approximated using indicators like ease of access and transfer. For instance, in shared vehicle service, a stated-preference study of inter-urban car sharing found that access time to parking locations is a key factor in choice, with elasticities showing a strong influence, especially for bus and carpool users. This supports treating short access times as a quantitative alternative for higher user satisfaction [105]. In a simulation, this can be implemented by recording the access time from the user’s origin to the nearest eligible station or parking location and reporting mean and percentile access times along with the share of trips meeting a policy threshold (e.g., access time ≤ 5 min) as a convenience-satisfaction indicator. Similarly, for multimodal platforms like MaaS, a study argues that effort and seamlessness are central to user experience. It proposes a simple way to quantify these factors by assigning each mode an inconvenience cost per unit time and mapping multimodal bundles along effort and seamlessness dimensions; for example, taxis are considered less inconvenient per minute than pooled ride share or public transport. [99]. In simulation, this can translate to an effort index built from in-vehicle time, waiting, walking, and transfer time, with mode-specific inconvenient weights that may vary with occupancy, paired with a seamlessness index based on transfer wait, transfer walk, and number of interchanges. Results can be reported as means, upper percentiles, and the share of itineraries that fall into low-effort, high-seamlessness categories, disaggregated by user group or corridor, as reproducible satisfaction alternatives. These indicators, while not direct measurements of satisfaction, provide valuable insights into factors that users generally associate with positive experiences.

By using alternative indicators, simulations can indirectly capture the influence of subjective factors on system performance. This approach offers a practical way to evaluate how advanced mobility systems may align with user expectations and preferences, enabling simulations to incorporate user-centered perspectives alongside operational indicators.

5. Conclusions

This study conducted a comprehensive review of public transport evaluation indicators, categorizing them as either direct or indirect to assess varied aspects of performance and impact. By examining existing evaluation indicators across dimensions like supply, demand, environmental, economic, and societal impact, the study highlights their continued applicability in the context of advanced mobility systems. Key findings indicate that while many existing indicators retain value, specific features of advanced mobility systems, such as flexible routing and scheduling, shared resources, and multimodal integration, necessitate adaptations to better capture their unique characteristics. The study emphasizes the need for complementary indicators to address these emerging requirements, ensuring a more holistic evaluation framework for advanced mobility systems.

Simulation modeling is a valuable tool for evaluating advanced mobility systems, enabling dynamic testing of system performance and user experience under varying scenarios. Planning indicators serve as input parameters that configure baseline conditions, while operating indicators act as performance benchmarks, revealing system efficiency. Indirect indicators, such as those for environmental and societal impact, extend the assessment to cover broader effects. Recognizing the challenges simulations face in capturing subjective factors, we explored alternative indicators as an indirect means of approximating user satisfaction, acceptance, and other qualitative dimensions, which are pivotal for user demand analysis.

In conclusion, as regions increasingly consider advanced mobility systems, our study underscores the need to adapt existing indicators to fully capture advanced mobility system potential and limitations. For future research, developing additional alternative indicators to reflect subjective aspects will be crucial for refining simulation-based evaluations and supporting advanced mobility system planning and integration into existing transit frameworks. In addition, future research should address the prioritization and weighting of indicators, as this is highly important for real-world decision-making. However, establishing such priorities requires additional questions to be addressed, such as clarifying stakeholder perspectives, defining policy objectives (efficiency, equity, sustainability), and selecting appropriate evaluation approaches. Prioritization may also require consideration of both compensatory and non-compensatory approaches. For compensatory approaches, multi-criteria decision analysis methods such as AHP [106] and TOPSIS [107] explicitly model trade-offs between indicators. For example, a user may prefer a bus to a taxi because it is cheaper but may instead choose a taxi if the time savings justify the higher fare. For non-compensatory approaches, such as elimination-by-aspects (EBA) [108], thresholds are applied rather than trade-offs. For example, governments may subsidize operators to lower fares, but if the subsidy level exceeds budget limits, the option becomes infeasible regardless of potential user benefits. These cases highlight that compensatory approaches allow disadvantages in one dimension to be balanced by advantages in another, whereas non-compensatory approaches exclude options outright once certain requirements are not satisfied. This illustrates that prioritization requires extensive deliberation and expert consultation and is therefore positioned as an important but separate research agenda beyond the scope of the present review. Finally, it should be noted that this paper is positioned as a theoretical exploration and framework development rather than an empirical validation. Future research should therefore complement this framework with real-world case studies and empirical testing, such as MaaS trials, DRT pilots, or micromobility programs, to demonstrate its practical applicability. Beyond indicator development, future work should focus on advancing the functionalities of simulation environments to ensure these indicators can be effectively implemented and analyzed under dynamic scenarios. Furthermore, enhancing demand analysis through the integration of big data sources, such as mobile application data and social media data, could enable a more dynamic and precise understanding of user needs, moving beyond traditional demographic-based approaches. Expanding the scope of analysis from regional systems to nationwide or even global contexts could also provide deeper insights into scalability and cross-regional coordination of advanced mobility systems. Together, these efforts will contribute to the creation of a more robust and adaptable evaluation framework capable of addressing the evolving challenges of modern transportation systems.