Service Quality Evaluation and Analysis of Autonomous-Rail Rapid Transit in Yibin City of China

Abstract

With the acceleration of urbanization, Autonomous-rail Rapid Transit (ART), as a new type of public transportation mode, plays an important role in alleviating traffic congestion and optimizing urban transportation structure. However, the operation of ART faces various problems, such as the route and station design problems considering passengers’ convenience and transferring efficiency, and there is a gap between passenger perception and expectation for the ART service quality. Therefore, it is crucial to comprehensively evaluate the service quality of ART, so as to improve passenger satisfaction and promote the sustainable development of ART. Taking Yibin ART as the research object, this study is based on the Service Quality (SERVQUAL) model, combined with the Analytic Hierarchy Process (AHP) and Fuzzy Comprehensive Evaluation (FCE), to analyze the service quality of Yibin ART. Firstly, a service quality evaluation indicator system for Yibin ART is constructed based on the extended SERVQUAL model that includes six dimensions of reliability, responsiveness, assurance, empathy, tangibility, and convenience, as well as 19 secondary indicators. Then, the research collects 110 valid samples through a questionnaire survey, and the rationality of the questionnaire is verified through reliability and validity analysis. Later, the weights of the indicators are calculated by AHP, and a comprehensive evaluation of Yibin ART service quality is conducted with the FCE method. Finally, based on the evaluation results, the study shows that the core indicators of the ART service quality are the service reliability and responsiveness, as well as the convenience; further, the results find the significant differences between participants’ perceptions and expectations for ART service quality, especially in the aspects of smooth driving, cleanliness, station location, ticket service and transferring, and the corresponding targeted strategies are proposed for improving the Yibin ART service quality. Additionally, future research will expand the sample and conduct in-depth research on passenger travel characteristics, carefully grasp the needs of passengers, continuously optimize operational service plans, and strive to improve the service level of ART.

1. Introduction

1.1. Research Background

With the acceleration of urbanization and the rapid growth of the urban population, transportation-related issues, such as traffic congestion and insufficient urban transportation capacity, are becoming increasingly prominent [1].

The urban rail transit system is a high-capacity, fast, safe, reliable, and organized transportation mode; it includes high-speed railways, subways, light rails, trams, monorails, and other urban rail networks. As a key solution to support mobility in high-density urban areas, it plays an essential role in daily transportation activities [2,3,4].

However, the complex interconnectivity, high costs, and limited flexibility of the urban rail transit have hindered their wide range of application and development [5]. So, currently, in first and second-tier cities of China, the urban rail transit systems have grown rapidly, but for China’s third-and fourth-tier cities, which mainly rely on bus transportation, the traffic congestion of major commercial areas is increasing [6]. Therefore, a brand-new urban rail transit mode is proposed, which is called Autonomous-rail Rapid Transit (ART) [5,6]. Due to the core features of ART (e.g., trackless operation, low cost, and high flexibility), ART is mainly used to solve the problem of public transportation congestion in third- and fourth-tier cities in China and enhance the image of cities [6,7].

ART is a new type of multiple-articulated rubber-tire transit, which uses intelligent perception, path tracking, and trajectory following control technologies to eliminate reliance on physical railway tracks, and the operation of the ART has comprehensively realized the adoption of technologies, such as power batteries, hydrogen energy, and wheel-edge motor drive [5]. The main features of ART include low cost, highly flexible features, and strong adaptability. Firstly, it does not require the laying of physical tracks, can utilize existing road resources, has a lower construction cost compared with a subway, and can complete line construction within one year. Secondly, ART occupies a smaller road width, provides flexible driving, and has better start-stop acceleration performance than traditional buses, making it suitable for dense road network environments. Additionally, ART can be compatible with existing public bus transportation systems, and can enjoy exclusive and shared road rights according to the road conditions, as well as the transportation volume is larger than that of buses, especially suitable for areas with low and medium volume demand, such as the third- and fourth-tier cities [5,6,7].

Because ART has the dual advantages of rail transit with punctuality, high-capacity, energy-saving, and environmental friendliness, as well as the flexibility and low overall cost of traditional bus operations [5,7], ART is gradually becoming a new choice for public transportation systems in small and medium-sized cities in China, such as the Zhuzhou Model Line, Suzhou ART T1 Line, Xi’an ART Line No. 1, Yibin ART T1 Line and T4 Line, etc. [6,7].

However, with the continuous advancement of urbanization, the construction and operations of urban rail transit have also encountered unprecedented development opportunities and various challenges [8]. Therefore, many scholars have conducted analyses of urban rail transit from various perspectives.

For example, Tang et al. [1] made great efforts to enhance the robustness of ART road object detection under different road conditions. Li et al. [2] proposed optimizing the resilience of urban rail systems with consideration of the impact of train delays and passenger congestion. Lu et al. [3] analyzed the coverage of spatial service, passenger service performance, and the operational efficiency of urban rail transit. Tang et al. [4] discussed the advanced technology in rail transit. Feng et al. [5] designed the ART architecture and operating principle. Zhou et al. [6] studied the problem of station location selection of ART. Yin et al. [7] studied the operation optimization problem of the ART cross-line. Rong [8] took the Yibin ART as an example to comprehensively analyze the main financial risks faced by the urban rail transit industry. Zou et al. [9] designed the station service system for ART. Chai et al. [10] developed an evaluation framework for the safety evaluation of urban rail transit operation. Bai et al. [11] established an evaluation indicator system to assess the development of urban rail transit across 27 Chinese cities.

As a public service system, the aim of these studies [1,2,3,4,5,6,7,8,9,10,11] is to improve the service quality of urban rail transit by various measures and promote the sustainable development of the urban rail transit system, because in urban rail transit systems, unexpected events, such as equipment breakdown, train delays, and crowded platforms, seriously affect the service quality to passengers [2]. Furthermore, in order to encourage the use of public transport, it is necessary to make the public transport attractive to passengers through regular service quality evaluation and modification [12].

Therefore, it is crucial to comprehensively evaluate the service quality of urban rail transit, which can identify the important factors affecting service quality and find the factors that make passengers dissatisfied, so as to improve the service quality of urban rail transit and provide important decision-making basis for transit operators and passengers, effectively promoting the stable and sustainable development of urban rail transit [11].

1.2. Literature Review

Regarding the research on the service quality of urban transportation, scholars worldwide have used different methods to analyze from different perspectives [12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36], and these studies also provide valuable references and ideas for our study.

For example, Sam et al. [12] analyzed the service quality expectations and perceptions of the core public bus transport users by using SERVQUAL (Service Quality) methodology with a paired-sample t-test and standard multiple regression techniques. The results found that reliability and responsiveness were the key points of the bus service quality. In order to improve the service quality of city public bus transportation, Duleba et al. [13] constructed an evaluation model from service quality, transport quality, and tractability. Firstly, they used the Analytic Hierarchy Process (AHP) to reveal a priority ranking of the elements of supply quality in Yurihonjo city of Japan. Later, Duleba and Moslem [14] employed Pareto optimality combined with AHP applications to determine the public preference on the importance of developing supply quality elements in local bus transportation service in Mersin, Turkey. Kutlu Gündoğdu et al. [15] extended the evaluation model in reference [13] and proposed a hybrid picture fuzzy analytic hierarchy process (PFAHP) and linear assignment method to assess public bus transportation service quality in Budapest, Hungary. Luke and Heyns [16] used the SERVQUAL model to measure the service quality of most public transportation modes by comparing the perceptions of passengers with their expectations, so as to determine the service gaps where the intervention and improvement were required. Farazi et al. [17] adopted a special clustering analysis together with a machine learning approach to investigate the impact of heterogeneity in user service quality perception on intercity train services in Bangladesh, thus providing valuable insights for sustainable development. Lu et al. [18] pointed out that more emphasis should be placed on the passenger experience and on enhancing the service quality of urban rail transit. They established a cascading failure model from urban rail transit network service capacity and service continuity, and employed the Monte Carlo simulation method to evaluate the disaster resilience capability of the Beijing subway network.

In addition, Halakoo et al. [19] focused on Taxi Khattee (TK) in Iran and assessed the indicators of service quality and their impact on the satisfaction of users in a shared taxi mode by using the Structural Equation Modeling (SEM) technique. Buran [20] developed a multiple linear regression model to analyze the impact factors on the satisfaction of passengers in public bus transportation. Muni et al. [21] used exploratory factor analysis (EFA) and confirmatory factor analysis (CFA), and the SEM approach to analyze the factors of service quality and their relationships with the satisfaction of passengers in rail freight transportation. Ong et al. [22] employed an integrated Social Exchange Theory (SET) and SERVQUAL model combined with causal analysis to analyze the influential factors affecting customer satisfaction, and SEM was employed to examine the service quality and passenger satisfaction of motorcycle taxi transportation in the Philippines. Aydin [23] proposed an evaluation framework to measure the service quality of rail transit systems via passenger satisfaction surveys, and a method that combined statistical analysis, fuzzy trapezoidal numbers, and Technique for Order Preference by Similarity to Ideal Solution (TOPSIS) was used to evaluate service quality levels for multiple periods. Watthanaklang et al. [24] analyzed the relationship between perceived accessibility of service quality and the intention of using public bus transportation systems with the methods of the modified SERVQUAL model, CFA and the SEM.

Gong et al. [25] pointed out that urban rail transit service quality assessments were crucial for transport authorities to measure passenger preferences and refine operational strategies, and they presented a comprehensive framework to efficiently classify and mine public opinion on urban rail transit services from social media platforms. Yuan et al. [26] used a passenger satisfaction index (PSI) model combined with a finite mixture partial least squares (FIMIX-PLS) technique and the importance-performance map analysis (IPMA) method to analyze the passenger satisfaction for air-rail integration services (ARIS). In order to improve the passenger satisfaction for the high-speed rail (HSR), Yang et al. [27] utilized a text mining of online reviews on social media and an improved Best-Worst Method under basic uncertain linguistic information environments (BULI-BWM) to identify the key passenger requirements and determine the importance of these requirements, further, an extended Quality Function Deployment (QFD) considering the uncertain information was applied to derive and prioritize improvement measures of service quality. In order to improve the service quality of Metro Rail Transit System (MRTS) and consequently attract more passengers, Mandhani et al. [28] studied the interrelationships among the service quality factors of MRTS by using an integrated Bayesian Networks (BN) and Partial Least Squares Structural Equation Modeling (PLS-SEM) approach. In order to understand and evaluate the service quality of public transport systems, Tumsekcali et al. [29] added two new indicators related to Industry 4.0 and the pandemic to extend the SERVQUAL model, and a novel methodology integrating AHP with Weighted Aggregated Sum Product Assessment (WASPAS) under an interval-valued intuitionistic fuzzy (IVIF) environment was employed to evaluate Istanbul public transport systems.

Because service quality is a key driving factor for passenger intention to reuse the urban rail transit. Wang and Shi [30] developed and used a multi-attribute decision-making model to evaluate the service quality of urban rail transit, and the assessment was mainly built on the interval-valued intuitionistic fuzzy value. Dziaduch and Peternek [31] employed the AHP method and the SUTI Indicator 4 (Public Transport Quality and Reliability) of the Sustainable Urban Transport Index (SUTI) to evaluate the service quality of public transport in Wrocław from the perspective of urban bus and tram users. Tiglao et al. [32] used EFA and SEM to evaluate the paratransit services, which referred to the informal mode operating in the urban transport system of Metro Manila, with the dominance of public utility jeepneys. In order to evaluate the service quality of urban rail transit, Lin [33] proposed a rough set theory applied to AHP, and the QFD method was applied to build a service quality evaluation indicator system of Fuzhou urban rail transit. Wang et al. [34] used natural language processing to quantify the rail transit service quality based on the social network data, and firstly, by using of the K-Means text clustering, passengers’ demands and service elements were transformed into evaluation indicators; then by use of the Term Frequency-Inverse Document Frequency (TF-IDF) method, the indicator weights were acquired; at last, the comprehensive score of service quality was obtained to analyze the evaluation results.

In order to improve the service quality of urban rail transit, He and Xu [35] constructed an evaluation indicator system of urban rail transit service quality considering passengers’ perceptions and demands, and a comprehensive evaluation method based on SEM and BN was developed. Tao and Xue [36] used the “5W1H” questionnaire, the KJ (Kawakita Jiro) method, the literature research, and other methods to construct a QFD quality house model for passengers, so as to build a service quality evaluation indicator system for urban rail transit, and Tianjin was taken as an example to conduct an empirical study.

Therefore, according to the literature review, regarding the research on the quality of urban transportation services, scholars worldwide have used different models and methods to analyze from different perspectives [12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36]. The main methods to establish the evaluation models and the methods used for assessing the service quality of various public transports in the references are shown in

Table 1. The main evaluation models and methods in the references.

Researcher	Time	Method for Establishing Evaluation Model	Evaluation Method	Object	Region
Sam et al. [12]	2018	SERVQUAL model	Paired-samples t-test, and standard multiple regression techniques	Public bus transport	Kumasi
Duleba et al. [13]	2012	Literature review and investigation studies	AHP	Public bus transportation	Yurihonjo city, Japan
Duleba and Moslem [14]	2019	Using reference [13]	Pareto optimality and AHP	Public bus transport	Mersin, Turkey
Kutlu Gündoğdu et al. [15]	2021	Using reference [13]	PFAHP-linear assignment method	Public bus transportation	Budapest, Hungary
Luke and Heyns [16]	2020	SERVQUAL model	Convenience sampling and t-tests	All the main public transport modes	Johannesburg, South Africa
Farazi et al. [17]	2022	Literature studies and expert counsel	Clustering approach with machine learning approach	Intercity train	Bangladesh
Lu et al. [18]	2025	Cascading failure model under Geometric Attack Model	Monte Carlo simulation method	Subway network	Beijing, China
Halakoo et al. [19]	2022	Literature studies	SEM, EFA, and CFA	Taxi Khattee	Tehran, Iran
Buran [20]	2025	Literature studies	Multiple linear regression model, and statistical analysis	Public bus transportation	Ten major cities worldwide
Muni et al. [21]	2024	Literature studies	SEM, EFA, and CFA	Rail freight transportation	Bangladesh
Ong et al. [22]	2024	SET and SERVQUAL model	Causal analysis and SEM	Motorcycle taxi transportation	Philippines
Aydin [23]	2017	Literature review and investigation studies	Statistical analysis, fuzzy trapezoidal numbers, and TOPSIS	Rail transit system	Istanbul
Watthanaklang et al. [24]	2024	Modified SERVQUAL model	CFA and SEM	Public bus transportation	Nakhon Ratchasima, Thailand
Gong et al. [25]	2024	Social media data (SMD)	A text classification and opinion mining algorithm	Subway	Ten Chinese cities
Yuan et al. [26]	2021	Literature studies	FIMIX-PLS technique and IPMA method	Air-rail integration services	Beijing-Tianjin-Hebei Metropolitan Region
Yang et al. [27]	2025	Social media online reviews, and KJ method	An improved BULI-BWM and an extended QFD	High-speed rail	China
Mandhani et al. [28]	2020	Literature studies	Integrated BN and PLS-SEM approach	Metro Rail Transit System	Delhi, India
Tumsekcali et al. [29]	2021	Extended SERVQUAL model	AHP integrated WASPAS under IVIF environment	All the main public transport modes	Istanbul
Wang and Shi [30]	2020	Literature studies	Interval-valued intuitionistic fuzzy model	Subway	Tianjin, China
Dziaduch and Peternek [31]	2024	Sustainable Urban Transport Index	AHP method	Public buses and trams	Wrocław, Poland
Tiglao et al. [32]	2020	Literature studies	EFA and SEM	Paratransit mode	Metro Manila
Lin [33]	2021	Literature studies	Rough set theory, AHP, and QFD	Subway	Fuzhou, China
Wang et al. [34]	2023	Social network data	Clustering approach together with TF-IDF method	Monorail and subway	Chongqing, China
He and Xu [35]	2023	Literature studies	SEM and BN	Subway	Kunming, China
Tao and Xue [36]	2019	“5W1H” questionnaire, KJ method, and literature studies	Independent point method, maximum correlation method, and QFD	Subway	Tianjin, China

.

From the literature review in Table 1, there are service quality evaluations for many types of public transports, such as bus [12,13,14,15,20,24,31], intercity train [17], subway [18,25,28,30,33,34,35,36], taxi [19], rail freight [21], motorcycle taxi [22], rail transit [23], air-rail integration service [26], high-speed rail [27], tram [31], paratransit [32], monorail [34], etc.

However, for the ART with the dual advantages of rail transportation and traditional bus operations, there is a lack of targeted research on the comprehensive evaluation of ART service quality. Therefore, the aim of the paper is to evaluate the service quality of the ART system, to understand the requirements of passengers, identify shortcomings of service, and provide an effective path for improving and enhancing the service quality of the ART system.

In addition, through literature research and analysis [12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36], based on the application of different public transportation modes in different countries and cities, due to the features of different research objects and the urban characteristics, it is necessary to establish different evaluation models and use different methods to carry out the evaluation from different perspectives. Among these evaluation models and methods, the main methods for establishing the evaluation indicator systems include SERVQUAL model [12,16,22,24,29], the literature studies [13,14,15,17,19,20,21,23,26,28,30,32,33,35,36], social network data [22,25,27,34], expert counsel [17], etc. The popular methods for evaluating the public transportation service quality are SEM [19,21,22,24,28,32,35], factor analysis, such as EFA and CFA [19,20,21,24,32], classification and clustering approach [17,25,34], multiple linear regression [12,20], multi-attribute decision-making method, such as AHP [13,14,15,29,31,33] and TOPSIS [23], fuzzy theory [15,23,29,30], QFD [27,33,36], BN [28,35], etc.

It can be seen that, in order to establish an evaluation model, the literature studies method is used very commonly [13,14,15,17,19,20,21,23,26,28,30,32,33,35,36], so as to comprehensively consider various factors that affect service quality. Similarly, social network data is used to collect the factors affecting service quality [22,25,27,34]. So, literature studies and social network data are always combined with other methods, such as SEM and factor analysis [19,20,21,28,32], to analyze the relationship between factors, as well as the effect on service quality. Then, for the other methods of establishing the evaluation models, SERVQUAL is a well-known and widely used method and quality tool for measuring service quality by investigating the gap between perceived quality and customer expectation [37,38]. The SERVQUAL model is also frequently used to evaluate the service quality of the public transportation system, which is modified and updated considering the requirements of the era [29]. So, in this paper, an extended SERVQUAL model is used to construct the evaluation indicator system for the ART service quality.

In addition, as a key subjective method with wide applicability in multiple criteria decision-making problems, AHP is frequently used for urban transportation problems, such as ranking of indicators for assessing public transport service quality and choosing modes of urban passenger transport [31]. Although AHP has great flexibility and wide applicability [39], in the decision-making process, due to the expression preferences from different perspectives or difficulty in making appropriate choices from a given numerical scale, expert knowledge may also be biased [15]. So, the AHP is always combined with fuzzy logic to weaken the uncertainty of subjective evaluation by experts [40], such as the references [15,29], which combined AHP and fuzzy theory to evaluate the service quality of public transport. So, in this paper, AHP combined with the fuzzy comprehensive evaluation (FCE) method is used to evaluate the service quality of ART, where the FCE is a fuzzy evaluation method based on fuzzy mathematics theory.

1.3. Research Content

Yibin city is a prefecture-level city under the jurisdiction of Sichuan Province, China. Due to the geographical factors in Yibin, it is difficult and costly to build a subway [41].

As the first city in southwestern China to introduce an Autonomous-rail Rapid Transit, Yibin has achieved significant results in the construction of the ART network [42]. The first line of Yibin ART, T1 Line, was officially opened for operation on December 5, 2019 [41]. Subsequently, T1 Line gradually optimized its operational services, and Yibin launched the construction of T2 Line and T4 Line, forming a backbone network architecture of “core area connectivity with industrial area coverage”, and achieving rapid connection between key areas such as high-speed rail hubs, old urban areas, and Sanjiang New Area. The Yibin ART appearance and the operation route are shown in Figure 1, and the three operating lines T1, T2, and T4 have a total operating mileage of 80 km, with a cumulative passenger capacity of over 22 million and a vehicle punctuality rate of over 98% [42]. The ART not only undertakes the urban transportation functions, but also deeply integrates the cultural characteristics of Yibin in its design, as shown in Figure 1, with the unique logos of “China’s Wine Capital” and “Ecological Bamboo City”.

Although Yibin ART can alleviate urban traffic pressure and improve urban operational efficiency, there are still some problems related to the services, such as passenger transferring [7] and station design [9]. So, it is necessary to assess and improve the service quality of ART.

The significance of conducting research using Yibin ART as a case study lies in two aspects. On the one hand, when the construction of the Yibin ART T1 Line started, it faced many doubts at that time, such as occupying too many ground transportation resources and affecting other vehicles. However, since the operation of the Yibin ART T1 Line, the ART has been able to highlight various adaptability, such as implementing independent road rights and shared road rights according to road conditions, making the ART efficient and convenient [41]. On the other hand, currently, Yibin ART has the longest operating mileage and the best operating effect in the world, forming the “Yibin Model” of a comprehensive solution for low and medium-volume transportation in the city, and providing a demonstration for other urban ART projects [42]. Similarly, taking Yibin ART as a case study, analyzing its service quality to identify the important factors and the weak points in the services, so as to make targeted strategies for improving the service quality of Yibin ART, which also provides reference value for improving the quality of ART services in other cities.

Therefore, the aim of the paper is to conduct an evaluation of the service quality of Yibin ART. Considering the issues with Yibin ART services, a comprehensive evaluation method is designed and used to model and evaluate the service quality of ART. Firstly, the general SERVQUAL framework is extended and used to model the evaluation indicator system. Then, the AHP combined with the FCE method is used to systematically evaluate and analyze the service quality of Yibin ART, so as to identify the important influencing factors on the service quality and find the gap between passenger satisfaction and expected service quality. Through evaluation, it can guide ART operators to improve service weaknesses in a targeted manner, enhance overall service quality, attract more passengers, and further promote the stable development of the ART, as well as the sustainable development of the green and low-carbon city.

The main contributions in this paper are as follows:

(1) An evaluation indicator system for Yibin ART service quality is constructed based on the extended SERVQUAL model.

(2) Based on the established evaluation indicator system, a comprehensive evaluation method is employed by integrating AHP with the FCE method to assess the ART service quality.

(3) Through the evaluation, the important influencing factors of service quality of Yibin ART are identified, and the weaknesses are found in the ART services, as well as the targeted improvement measures are proposed, so as to improve the service quality of Yibin ART and guide the sustainable development of ART.

The research content of the paper is that, firstly, Section 1 is the literature review about the study on the public transportation service quality, and Section 2 proposes an evaluation indicator system for the Yibin ART service quality based on the SERVQUAL model. Then, Section 3 uses the AHP and FCE method to determine the weights of service quality indicators and conduct the comprehensive evaluation. Later, in Section 4, an empirical analysis of the evaluation of Yibin ART service quality is carried out based on the established evaluation indicator system and the comprehensive evaluation method. Finally, Section 5 summarizes the conclusions and proposes future research.

2. Evaluation Indicator System Model for Yibin ART Service Quality

In order to evaluate the service quality of Yibin ART, the commonly used SERVQUAL model framework is employed to establish the evaluation indicator system. The SERVQUAL model is a useful tool for analyzing the gap between customers’ expectations and their perceptions by using various criteria/attributes, and the frequently used criteria of the SERVQUAL model include tangibility, responsiveness, reliability, assurance, and empathy, which have been repeatedly validated to have good reliability and validity [43,44,45]. However, with the development of the era and the changes and increases in customer needs, when using the SERVQUAL model, it is often updated and modified considering the requirements of the era [29]. So, when constructing an evaluation model for the urban rail transit service quality, the SERVQUAL model is always extended.

Therefore, except for the five basic attributes of the SERVQUAL model, which are reliability, responsiveness, assurance, empathy, and tangibility, the convenience is also included to evaluate the service quality of Yibin ART. Convenience means that the good design and reasonable layout of the urban rail transit service system, including the convenience and intelligence of entry and exit, ticketing, and transfer, can help improve the efficiency of passenger travel [33].

The indicators related to the convenience are also mentioned and used in many references, such as references [13,14,15,16,17,20,23,24,26,27,28,30,31,32,33,34,35,36], especially the evaluation of urban rail transit in China [26,27,30,33,34,35,36]. In addition, considering the issues of the Yibin ART reflected in the references [7,8,9], and the voice of the masses from Sichuan Province Online Mass Work Platform [46,47], many problems are related to the passenger convenience needs, such as information accessibility, convenience of transfer, clarity of directional signs, etc. [7,9,46,47].

Therefore, based on the SERVQUAL model, this paper adds the attribute of “convenience” to form a six-dimensional evaluation indicator system for the service quality of Yibin ART, including reliability, responsiveness, assurance, empathy, tangibility, and convenience. For the selection and determination of the secondary indicators, this paper mainly refers to the relevant references [12,13,14,15,16,17,21,23,24,25,26,27,28,29,30,31,32,33,34,35,36,48], and the selection criteria are drawn on the existing research results and expert opinions to select specific indicators that are widely used and closely related to the service quality of urban rail transit and public bus transportation.

Simultaneously, considering the actual needs and expectations of passengers in Yibin city [7,9,46,47], key elements that can directly affect their travel experience and satisfaction are also selected as the secondary indicators. Ultimately, a six-dimensional indicator system for Yibin ART service quality is formed as shown in

Table 2. Six-dimensional evaluation indicator system for Yibin ART service quality.

First-Level Indicator	Second-Level Indicator	Description	Reference
A1-Reliability	B1-Punctuality	The ART usually arrives on time.	[12,16,17,23,24,26,27,28,29,30,31,33,35]
	B2-Operation schedule	ART follows a schedule.	[12,21,24,26,27,29,31,33,34,35]
	B3-Smooth driving	The ART drives smoothly, does not rush or brake suddenly.	[16,24,33,34,35,36]
A2-Responsiveness	B4-Passenger needs	The requests of passengers are promptly responded to.	[12,21,24,26,27,29,30,35]
	B5-Route information	Real-time updates of station information on the ART APP and electronic screen.	[12,13,14,15,21,23,26,28,29,31,34]
	B6-Adequate transport	Adequate transport is provided for services during peak hours, on weekends, and on public holidays.	[16,24,34]
A3-Assurance	B7-Staff attitude	Staff are always polite and smart.	[12,17,23,24,25,27,28,29,33,34,35,36]
	B8-Guidance signs	The guidance signs in the ART station are clear.	[26,33,34,35,36]
	B9-Travel safety	Feeling in safe, such as the passenger belongings are secured.	[12,13,14,15,16,17,21,23,24,25,26,28,29,30,31,33,34,35,36]
A4-Empathy	B10-Accessibility service	Providing barrier-free facilities, such as the suitability of stations for disabled passengers.	[29,48]
	B11-Individual service	Individual attention to passengers, such as seat availability for persons with disability, women, and senior citizens inside ART.	[28,29]
	B12-Passenger feedback	Passenger feedback channels, such as online platforms, are unobstructed.	[29,35]
A5-Tangibility	B13-Cleanliness	Cleanliness of inside the ART and stations.	[12,16,17,23,24,28,29,30,31,34,35,36]
	B14-Maintenance	The ART facilities, such as seats and the air-conditioning system, are well-maintained, making passengers feel comfortable.	[12,16,17,23,24,25,26,27,29,30,31,33,34,35,36]
	B15-Lighting system of station	The lighting system is sufficient in station, providing a high sense of safety at night.	[23,28,31,36]
	B16-Staff appearance	The staff get dressed in neat and clean clothes.	[12,24,27,29,35]
A6-Convenience	B17-Station location	The location of stations is suitable and convenient.	[16,24,27,31]
	B18-Ticket service	ART tickets are easily accessible.	[12,17,23,24,26,27,28,29,30,33,34,35,36]
	B19-Transferring	Desirable route, and transferring with other transportation modes is convenient.	[24,26,28,33,35,36]

.

In Table 2, for the first-level, the indicators include A1-Reliability, A2-Responsiveness, A3-Assurance, A4-Empathy, A5-Tangibility, and A6-Convenience. A1-A5 indicators are the core basic attributes of the SERVQUAL framework, and A6-Convenience is an extended attribute considering the service needs of Yibin ART. In addition, considering the ART with the dual advantages of the rail transportation and the traditional bus operations [5,7], the widely used bottom indicators in the relevant references about the urban rail transit and public bus transportation are selected [12,13,14,15,16,17,21,23,24,25,26,27,28,29,30,31,33,34,35,36]. Simultaneously, based on the existing service quality issues in the operation of Yibin ART and the requirements of the Ministry of Transport [7,9,46,47,48], there are 19 second-level indicators selected corresponding to these six dimensions. The detailed description of the indicators is as follows.

(1) A1-Reliability, it is the ability of ART to perform the promised service dependably and accurately. There are three second-level indicators, including B1-Punctuality, B2-Operation schedule, and B3-Smooth driving, to evaluate the punctuality and operation quality of Yibin ART.

(2) A2-Responsiveness, it means that the ART staff should respond promptly to passenger needs and quickly answer inquiries, as well as reasonably arrange vehicle operation intervals and reduce passenger waiting time. There are three second-level indicators, including B4-Passenger needs, B5-Route information, and B6-Adequate transport.

(3) A3-Assurance, it means the professional knowledge and courtesy of the ART staff, and their ability to convey trust and confidence, to provide accurate information and high-quality services to passengers. There are three second-level indicators, including B7-Staff attitude, B8-Guidance signs, and B9-Travel safety.

(4) A4-Empathy, it means that the ART staff consider issues from the perspective of passengers, pay attention to their special needs, and provide personalized services such as barrier-free access and priority seating. There are three second-level indicators, including B10-Accessibility service, B11-Individual service, and B12-Passenger feedback.

(5) A5-Tangibility refers to the physical appearance of the vehicles, equipment, staff, and other infrastructure, which will affect passengers’ first impression and enhance their trust in the service. There are four second-level indicators, including B13-Cleanliness, B14-Maintenance, B15-Lighting system of station, and B16-Staff appearance.

(6) A6-Convenience, it means that the ART system provides the passengers with various conveniences to enhance the efficiency of the travel, such as the convenience for entry and exit, ticketing, and transfer. There are three second-level indicators, including B17-Station location, B18-Ticket service, and B19-Transferring.

Because the application of SERVQUAL model involves the use of structured questionnaires, and these questionnaires are used to ask customers about their expectations and perceptions on services [38]. So, in this paper, with a Likert 5-point scale, a questionnaire is designed about the satisfaction and importance of these 19 indicators from the perspectives of passengers to evaluate the ART service quality. Satisfaction reflects the true feelings and evaluations of passengers when they are using the ART, while importance reflects the expectations of passengers for the quality of various aspects of ART services in the future.

3. Evaluation Method Design for Yibin ART Service Quality

After constructing a six-dimensional evaluation system based on the extended SERVQUAL model, the evaluation analysis of Yibin ART service quality can be further carried out.

The FCE method is a decision-making method based on fuzzy mathematics theory, and can effectively handle some difficult-to-quantify problems, such as sentiment analysis, expert systems, fuzzy control, and other fields [49].

The reason for using the FCE method to evaluate the ART service quality is that it can comprehensively consider multiple evaluation indicators, handle the ambiguity and uncertainty between evaluation indicators, and obtain comprehensive evaluation results, providing more flexible and robust decision-making solutions [49,50].

The steps of FCE to evaluate the service quality of Yibin ART mainly include calculating the weights of indicators in the evaluation system, constructing an evaluation set of these indicators and the membership degree matrix of the evaluation set, and calculating fuzzy comprehensive score with fuzzy synthesis operation. The specific evaluation steps are as follows.

3.1. Calculating the Indicator Weights by AHP

Due to the different roles played by different service quality evaluation indicators in the assessment process, it is reasonable to give different weights to different evaluation indicators.

There are two kinds of methods for determining the weights of indicators, including the subjective weight method and the objective weight method. Subjective weight method, such as AHP, is based on the expert experience and simple to use, but easily influenced by personal preferences; objective weight method, such as entropy weight method, is relatively rigorous but has a high requirement for data quality; the choice of appropriate method depends on multiple factors, such as the specific decision situation, available data, and time resources [49].

For the decision-making technique, AHP was first published by Saaty [51], and Saaty also constructed the Analytic Network Process (ANP), while AHP is more popular than ANP due to the complexity in the application of ANP [14]. Because in AHP, the importance of each indicator is defined based on the pairwise comparisons [52], while ANP needs not only the pairwise comparisons of the indicators from the aspect of importance but also from the aspect of the impact on each other, and ANP is usually used for those complex decisions, in which the criteria are not only hierarchically connected [14].

Because there is only a hierarchical relationship between the upper and lower levels in the evaluation indicator system for Yibin ART service quality, and the evaluation for the service quality indicators is based on passengers’ and other stakeholders’ own feelings, in this paper, as the widely used subjective weight method, AHP is employed to determine the weights of evaluation indicators. The specific process of AHP is below.

(1) Establishing judgment matrix

In AHP, the importance of an indicator is expressed in the relative significance coefficients [53]. So, the evaluation indicators need to be compared in pairs to construct the expert judgment matrices. In this paper, ten experts from passengers and operators are invited to express their subjective feelings about the significance of the indicators. Although the number of participants is evidently not statistically representative, all the multi-attribute decision-making techniques, such as AHP, can provide deeper insights based on pairwise comparisons than simple statistical surveys [14].

If on the same level, there are n indicators with the same affiliation, and an expert judgment matrix A can be obtained as follows:

$$A=\left[\begin{array}{cccc}{a}_{11}& a_{12}& \dots & a_{1n}\\ a_{21}& a_{22}& \dots & a_{2n}\\ ⋮& ⋮& a_{ij}& ⋮\\ a_{n1}& a_{n2}& \dots & a_{nn}\end{array}\right]$$

(1)

In matrix A, a_ij (i, j = 1, 2, …, n) reflects the importance of indicator i compared to indicator j, while a_ji represents the importance of indicator j relative to indicator i.

The methods for determining the indicator importance of pairwise comparisons include 1–5 scale scoring criteria [54], 1–9 scale scoring criteria [51,52], and other scales such as scale of interval valued fuzzy number [15,29], and for the evaluation, AHP recommends the 1–9 scale which is called Saaty-scale [13,51], and it is also the mainstream method used in the service quality evaluation of various public transports [13,14,31,33].

So, in this paper, the 1–9 scale scoring criteria with the discrete set {9, 8, 7, 6, 5, 4, 3, 2, 1, 1/2, 1/3, 1/4, 1/5, 1/6, 1/7, 1/8, 1/9} are employed for the quantitative pairwise comparison of the importance degree between indicators [51]. If a_ij = 1, the indicator i is as important as indicator j; if a_ij = 3, the indicator i is slightly more important than indicator j; if a_ij = 5, the indicator i is significantly more important than indicator j; if a_ij = 7, the indicator i is much more important than indicator j; if a_ij = 9, the indicator i is extremely important than indicator j; if a_ij = 2, 4, 6, 8, it is the median of the above judgments; and a_ji = 1/a_ij, it is the importance of indicator j to indicator i.

(2) Consistency checking

In order to ensure the rationality of the allocation of indicator weights, consistency checking of the expert judgment matrix is required. Consistency Ratio (CR) is used as a measure of consistency [52]. The calculation of the CR value is as follows.

First of all, the Geometric Mean Method is employed to calculate the indicator weights of the judgment matrix A by the formula below:

$$w_i={\displaystyle \frac{{({\Pi }_{j=1}^n{a}_{ij})}^{1/n}}{{\displaystyle {\sum }_{i=1}^n{({\Pi }_{j=1}^n{a}_{ij})}^{1/n}}}},i=1,2,\dots ,n$$

(2)

where w_i is the weight of indicator i.

Then, calculating the maximum eigenvalue λ_max by the following formula:

$${\lambda }_{\max }={\displaystyle {\sum }_{i=1}^n\frac{{[AW]}_i}{n{w}_i}},i=1,2,\dots ,n$$

(3)

where W is the weight set of indicators, and [AW]_i is the i-th (i = 1, 2, …, n) component of matrix [AW].

Finally, calculating the CR value by the following formula:

$$CR={\displaystyle \frac{CI}{RI}}={\displaystyle \frac{{\lambda }_{\max }-n}{(n-1)RI}}$$

(4)

where CI is the consistency index, and RI is a random mean consistency index. This study adopts the standard RI values suggested by Saaty: n = 3, RI = 0.52; n = 4, RI = 0.89; n = 5, RI = 1.11; n = 6, RI = 1.25; n = 7, RI = 1.35 [52]. If the CR value is less than 0.1, the checking of consistency of the expert judgment matrix is passed and the indicator weights are acceptable; otherwise, the expert needs to re-evaluate, and the checking process is repeated until the CR value is less than 0.1.

3.2. Constructing Evaluation Set and Membership Degree Matrix

(1) Constructing evaluation set

In order to gain a specific understanding of the passengers’ usage experience and future expectation for Yibin ART service quality, an online survey questionnaire is designed. The survey mainly targets the residents of Yibin city. The questionnaire is divided into two parts. The first part is about the basic information of the respondents, including gender, age, frequency, and reasons for taking the ART. The second part is divided into two aspects, including satisfaction survey and importance survey, to collect passengers’ satisfaction and importance evaluation for the 19 bottom indicators of Yibin ART service quality. Satisfaction reflects the true experience of passengers, and importance reflects the expectation of passengers for the Yibin ART service quality in the future.

The satisfaction grade and importance grade are both designed using a Likert 5-point scale. The satisfaction evaluation set = (very satisfied, satisfied, general satisfied, dissatisfied, very dissatisfied), and each satisfaction level corresponds to a different score on the Likert 5-point scale = (5, 4, 3, 2, 1). The importance evaluation set = (very important, important, general important, unimportant, very unimportant) = (5, 4, 3, 2, 1). After the questionnaire items of the 19 indicators are answered by respondents based on their actual feelings, the evaluation sets of passenger satisfaction and importance can be obtained.

(2) Constructing the membership degree matrix

When using the Likert 5-point scale to evaluate the satisfaction and importance degree of a single indicator separately, the membership degree can be calculated by collecting a large amount of questionnaire data and counting the frequency of elements belonging to a set, so as to obtain the membership degree matrix of the evaluation set for Yibin ART service quality, and the membership degree matrix is as follows:

$$R=\left(\begin{array}{ccc}{r}_{11}& \dots & r_{15}\\ ⋮& r_{ik}& ⋮\\ r_{n1}& \dots & r_{n5}\end{array}\right)$$

(5)

where r_ik represents the membership degree of the k-th (k = 1, 2, …, 5) evaluation grade made by passengers on the i-th (i = 1, 2, …, n, and n = 19) indicator. The value of membership degree is usually between 0 and 1, representing the varying degree to which the element belongs to the fuzzy set. If the value of membership degree is 0, it indicates that the element does not belong to the fuzzy set at all, and a value of 1 represents full membership in the fuzzy set [49].

Based on the data obtained from the questionnaires, two membership degree matrices of passenger satisfaction and importance for the Yibin ART service quality can be constructed.

3.3. Fuzzy Comprehensive Evaluation with Fuzzy Synthesis Operation

(1) Fuzzy synthesis operation

Fuzzy synthesis operation is the synthesis of the weight set obtained from AHP and the membership degree matrix to achieve the comprehensive evaluation matrix. This evaluation matrix reflects the comprehensive membership degree of the evaluation object in various evaluation grades. The calculation formula of the comprehensive evaluation matrix is as follows:

$$B=[b_1,\dots ,b_m,\dots ,b_6],\mathrm{and} b_m=W_m{R}_m,(m=1,2,\dots ,6)$$

(6)

where B is the comprehensive evaluation matrix, b_m is the membership degree vector corresponding to the first-level indicator m, W_m is the weight set of the second-level indicators corresponding to the first-level indicator m, and R_m is the membership degree matrix of these second-level indicators. For the evaluation indicator system of the Yibin ART service quality, there are six first-level indicators, so here, m =1, 2, …,6.

(2) Fuzzy comprehensive evaluation

Based on the scores of passenger satisfaction and importance corresponding to the Likert 5-point scale (5, 4, 3, 2, 1), the fuzzy evaluation scores of the first-level indicators on passenger satisfaction and importance for the Yibin ART service quality are obtained by the formula below:

$$S=B{V}^T$$

(7)

where S is the fuzzy evaluation scores of the first-level indicators, $V^T={(5,4,3,2,1)}^T$, and B is the comprehensive evaluation matrix.

Then, the comprehensive evaluation matrix B is combined with the weight set W of the first-level indicators to obtain the fuzzy comprehensive weights as $W_C=BW$. Further, the fuzzy comprehensive weights Wc are integrated with the five-level evaluation set $V^T={(5,4,3,2,1)}^T$, the comprehensive evaluation score of the Yibin ART service quality can be obtained as $S_C=W_C{V}^T$.

According to the comprehensive evaluation score, the grade to which the evaluation level belongs can be determined. Finally, by comparing the gap between the passenger satisfaction and the expected service quality, the service indicators can be identified that passengers are dissatisfied, so as to make targeted improvements to promote the sustainable development of Yibin ART.

Therefore, based on the description above, the service quality evaluation process for Yibin ART is shown in Figure 2, and the specific description is below.

Firstly, according to the established evaluation indicator system based on the extended SERVQUAL model, pairwise comparison by using the 1–9 scale scoring criteria is carried out to construct the expert judgment matrix, and the CR value is calculated to check the consistency of each expert judgment matrix. If the CR value is greater than 0.1, then the judgment matrix is readjusted until it meets the consistency requirement that the value of CR is less than 0.1. Next, using the Geometric Mean Method to calculate the weights of indicators in the AHP, so as to obtain the weight set of indicators. Then, using the Likert 5-point scale to evaluate the passenger satisfaction and importance, we establish the evaluation sets of passenger satisfaction and importance. Further, with the evaluation sets, the membership degree of each indicator can be calculated, and the membership degree matrix can be constructed. Later, through the fuzzy synthesis of the weight set obtained from AHP and the membership degree matrix, the comprehensive evaluation matrix is obtained. Finally, the fuzzy evaluation scores are calculated based on the comprehensive evaluation matrix and the score distribution of the Likert 5-point scale, and the results analysis can be conducted based on the evaluation scores.

4. Yibin ART Service Quality Evaluation and Results Analysis

4.1. Data Acquisition and Statistical Analysis

In this study, data acquisition is performed by a questionnaire survey. The questionnaire is designed based on the six-dimensional evaluation indicator system of Yibin ART service quality, and the questionnaire is answered by passengers with the Yibin ART riding experience to express their satisfaction and expectation for the ART service quality.

The questionnaire adopts the Likert 5-point scale, requiring the surveyed passengers to rate their satisfaction with Yibin ART services based on their current or previous travel experience, and give the importance of the corresponding indicators to reflect their expectation for Yibin ART services. The score design ranges from 1 to 5 points, corresponding to the five levels of satisfaction from “very dissatisfied” to “very satisfied”, and five levels of importance from “very unimportant” to “very important”.

This survey questionnaire was distributed through a professional platform named “Wenjuanxing”. “Wenjuanxing” is a professional online survey platform that provides questionnaire design, data collection, and statistical analysis services. It is widely used in enterprises, universities, and personal fields, and supports functions such as customer satisfaction surveys and academic research.

The questionnaire investigations were performed between March and May 2025. A total of 132 questionnaires were collected, and the received questionnaires were screened, and 22 invalid questionnaires that were not answered according to the regulations or answered incompletely were removed. Finally, 110 valid questionnaires were collected, with a validity rate of 83.33%. On the basis of the statistical techniques employed in this study (i.e., reliability and validity analysis), the sample is adequate to establish statistical significance [12,22].

The sample characteristics are shown in

Table 3. The sample characteristics.

Characteristic	Item	Quantity	Percentage (%)
Gender	Female	50	45.5
	Male	60	54.5
Age	Less than 18 years old	12	10.9
	18–35 years old	54	49.1
	36–55 years old	39	35.5
	More than 55 years old	5	4.5
Frequency of using ART per week	Once a week	19	17.3
	2–4 times per week	46	41.8
	More than 4 times per week	45	40.9
Reason for using ART	Convenience and saving time	62	-
	Punctuality	60	-
	Affordable prices and low travel costs	68	-
	Comfortable riding environment	43	-
	Safety	36	-

. There are 50 females (45.5%) and 60 males (54.5%), of which 10.9% are less than 18 years old, 49.1% are between 18 and 35 years old, 35.5% are between 36 and 55 years old, and 4.5% are more than 55 years old. There are 17.3% using ART with the frequency of once a week, 41.8% with the frequency of 2–4 times per week, and 40.9% with the frequency of more than 4 times per week. The reason for using ART is a multiple-choice question, and the main reasons for using ART include affordable prices and low travel costs, convenience and saving time, and punctuality.

Additionally, the statistical dataset of 110 questionnaires is shown in

Table 4. The statistical dataset.

Question	Very Satisfied/ Important (%)	Satisfied/Important (%)	General Satisfied/Important (%)	Dissatisfied/Unimportant (%)	Very Dissatisfied/Unimportant (%)
B1. The ART usually arrives on time.	0.23/0.31	0.29/0.39	0.27/0.23	0.18/0.06	0.03/0.01
B2. ART follows a schedule.	0.24/0.32	0.35/0.41	0.16/0.21	0.20/0.04	0.05/0.03
B3. The ART drives smoothly, does not rush or brake suddenly.	0.17/0.39	0.35/0.40	0.29/0.15	0.12/0.05	0.06/0.01
B4. The requests of passengers are promptly responded to.	0.19/0.35	0.35/0.28	0.18/0.23	0.23/0.14	0.05/0.01
B5. Real-time updates of station information on the ART APP and electronic screen.	0.23/0.23	0.27/0.45	0.25/0.20	0.16/0.09	0.08/0.04
B6. Adequate transport is provided for services during peak hours, on weekends, and on public holidays.	0.22/0.28	0.35/0.38	0.16/0.17	0.19/0.16	0.08/0
B7. Staff are always polite and smart.	0.25/0.22	0.40/0.30	0.18/0.28	0.15/0.15	0.03/0.05
B8. The guidance signs in the ART station are clear.	0.25/0.17	0.34/0.36	0.18/0.24	0.18/0.18	0.05/0.05
B9. Feeling in safe, such as the passenger belongings are secured.	0.23/0.22	0.34/0.38	0.31/0.21	0.10/0.12	0.02/0.07
B10. Providing barrier-free facilities, such as the suitability of stations for disabled passengers.	0.30/0.22	0.36/0.35	0.22/0.26	0.08/0.14	0.04/0.03
B11. Individual attention to passenger, such as seat availability for persons with disability, women and senior citizens inside ART.	0.28/0.17	0.30/0.39	0.28/0.29	0.10/0.14	0.04/0.01
B12. Passenger feedback channels, such as online platforms, are unobstructed.	0.21/0.26	0.42/0.27	0.25/0.30	0.05/0.15	0.07/0.02
B13. Cleanliness of inside the ART and stations.	0.15/0.37	0.41/0.44	0.26/0.12	0.14/0.06	0.04/0.01
B14. The ART facilities, such as seats and air-conditioning system, are well maintained, making passengers feel comfortable.	0.22/0.31	0.35/0.45	0.25/0.21	0.15/0.02	0.03/0.01
B15. The lighting system is sufficient in station, providing a high sense of safety at night.	0.19/0.33	0.39/0.42	0.29/0.21	0.08/0.04	0.05/0.01
B16. The staff get dressed in neat and clean clothes.	0.18/0.29	0.35/0.48	0.30/0.19	0.14/0.01	0.03/0.03
B17. The location of stations is suitable and convenient.	0.15/0.35	0.35/0.43	0.26/0.14	0.20/0.05	0.05/0.03
B18. ART tickets are easily accessible.	0.20/0.35	0.29/0.39	0.23/0.17	0.25/0.08	0.03/0
B19. Desirable route, and transferring with other transportation modes is convenient.	0.18/0.35	0.31/0.40	0.29/0.17	0.19/0.08	0.03/0

; the questions in Table 4 from B1–B19 correspond to the description of the indicators B1–B19. By using the Likert 5-point scale, the percentages of satisfaction and importance are calculated and correspond to the respondents’ evaluations of ART service satisfaction and expectation, respectively. The dataset is the foundation for analyzing and evaluating the quality of ART services.

Based on the statistical dataset in Table 4, the Cronbach’s alpha (CA) coefficient is the most commonly used index to assess the internal consistency reliability. A higher value of Cronbach’s alpha means better internal consistency reliability. By SPSS, version 22.0, the Cronbach’s alpha value obtained was 0.913, and it is greater than 0.70, which indicates that the questionnaire can be utilized for analysis [22,50].

Further, in order to assess the validity of the questionnaire, it is crucial to calculate the Kaiser-Meyer-Olkin (KMO) test and Bartlett’s Test of Sphericity. The KMO index was calculated to be 0.775, which is greater than 0.5, indicating that the sample size is adequate for factor analysis [21,50]. Bartlett’s Test of Sphericity was found to be statistically significant (χ² = 2090.218, df = 703, p-value = 0.000), where p-value < 0.05, indicating that it was possible to conduct a meaningful factor analysis using the data [21].

4.2. Service Quality Evaluation

In this section, aimed at improving the service quality of Yibin ART, according to the evaluation process, after constructing the hierarchical structure of the evaluation indicator system for Yibin ART service quality, the next step is to establish the indicator weight set by AHP. So, firstly, this paper explains the calculation process of the weights of indicators by AHP. Later, the evaluation steps of FCE are shown via an example. Lastly, the results analysis for the service quality of Yibin ART is presented and discussed.

4.2.1. Weights of Indicators

In AHP, the weights of indicators are calculated based on the expert judgment matrix. The reference [29] described forming an expert group to obtain opinions on the pairwise comparisons of indicators in the public transportation service quality evaluation, and there were five experts, including two academicians working in the related field, two managers from public and private transportation sectors, and one expert was an employee of the surveyed subject.

In this study, referring to the reference [29], there are three different types of evaluators in the expert group, including academics from different departments of universities and relevant personnel from the public transport management department. There are a total of ten experts, among them four experts are academics working in the related field, three managers come from Yibin ART transport management department, and the other three experts are the staff of Yibin ART having over 3 years of work experience. All experts have the same importance weights, that is, there is no difference between experts.

The ten experts are invited to give their opinions on the relative importance of indicators in pairwise comparisons. The pairwise comparisons using a 1–9 scale scoring criteria to construct the expert judgment matrices of the ten experts, and the consistency ratio CR of each judgment matrix is computed by Equation (4). The consistency ratio of each judgment matrix is less than 0.1, which means that all the judgment matrices are tolerably consistent. Then, judgment matrices taken from different experts are aggregated by the Geometric Mean Method, and by using Equation (2), the indicator relative weights can be obtained. Since the relative weights of the first-level indicators are also the global weights, further, by integrating the relative weights of the indicators in the first-level with the second-level, the global weights of the second-level indicators can be obtained as shown in

Table 5. The indicator weights.

First-Level Indicator	Relative Weight W1	Second-Level Indicator	Relative Weight W2	Global Weight W1 ∗ W2
A1-Reliability	0.2913	B1-Punctuality	0.3248	0.0946
		B2-Operation schedule	0.3470	0.1011
		B3-Smooth driving	0.3282	0.0956
A2-Responsiveness	0.2529	B4-Passenger needs	0.2665	0.0674
		B5-Route information	0.0980	0.0248
		B6-Adequate transport	0.6355	0.1607
A3-Assurance	0.0666	B7-Staff attitude	0.4982	0.0332
		B8-Guidance signs	0.1793	0.0119
		B9-Travel safety	0.3225	0.0215
A4-Empathy	0.0600	B10-Accessibility service	0.3625	0.0218
		B11-Individual service	0.4657	0.0279
		B12-Passenger feedback	0.1718	0.0103
A5-Tangibility	0.1433	B13-Cleanliness	0.2191	0.0314
		B14-Maintenance	0.2213	0.0317
		B15-Lighting system of station	0.0992	0.0142
		B16-Staff appearance	0.4603	0.0660
A6-Convenience	0.1859	B17-Station location	0.2064	0.0384
		B18-Ticket service	0.5666	0.1053
		B19-Transferring	0.2270	0.0422

. Through the global weights, the importance of each indicator can be analyzed.

According to Table 5, for the importance of the first-level indicators of the Yibin ART service quality, A1-Reliability (0.2913) and A2-Responsiveness (0.2529) are the core attributes, with a total weight of over 50%, reflecting the high concern for operational stability and service efficiency. A6-Convenience (0.1859) and A5-Tangibility (0.1433) are of secondary importance, reflecting the emphasis on travel experience and facility environment. The weights of A3-Assurance (0.0666) and A4-Empathy (0.0600) are relatively low, indicating that the importance of the basic service guarantee is relatively weak.

Further, in order to compare the importance of indicators more intuitively, the weight values of the second-level indicators are shown in Figure 3, and the specific analysis is as follows.

From Figure 3, it can be seen that, for the second-level indicators of the Yibin ART service quality, the important indicators include B6-Adequate transport (0.1607), B18-Ticket service (0.1053), B2-Operation schedule (0.1011), B3-Smooth driving (0.0956), and B1-Punctuality (0.0946), and these indicators further illustrate the importance of operational efficiency and convenience.

The relatively unimportant indicators include B12-Passenger feedback (0.0103), B8-Guidance signs (0.0119), and B15-Lighting system of station (0.0142). The possible reason is that the construction of the station is on the ground and has few intersecting routes, so it is easy to get on and off the vehicles, with fewer needs for guidance signs and staff. In addition, the ART stations are smaller and simpler compared to the subways, making it easier for passengers to enter and exit, so passengers place greater emphasis on driving efficiency.

4.2.2. Application of FCE

According to the steps of FCE for evaluating the service quality of Yibin ART, after establishing the indicator weight set by AHP, the membership degree matrices of the passenger satisfaction and importance for the Yibin ART service quality need to be calculated, respectively.

Based on the data obtained from the questionnaires about passenger satisfaction and importance evaluation on the service quality of Yibin ART, by counting the frequency of different evaluation grades belonging to each five-level set of passenger satisfaction and importance, the membership degree of each indicator in different levels can be calculated. Finally, two membership degree matrices of passenger satisfaction and importance for Yibin ART service quality can be constructed as $R_{19\times 5}$, and the results are shown in

Table 6. Calculation results of passenger satisfaction membership degree.

First-Level Indicator	Second-Level Indicator	Very Satisfied	Satisfied	General Satisfied	Dissatisfied	Very Dissatisfied
A1-Reliability	B1-Punctuality	0.23	0.29	0.27	0.18	0.03
	B2-Operation schedule	0.24	0.35	0.16	0.20	0.05
	B3-Smooth driving	0.17	0.35	0.29	0.12	0.06
A2-Responsiveness	B4-Passenger needs	0.19	0.35	0.18	0.23	0.05
	B5-Route information	0.23	0.27	0.25	0.16	0.08
	B6-Adequate transport	0.22	0.35	0.16	0.19	0.08
A3-Assurance	B7-Staff attitude	0.25	0.40	0.18	0.15	0.03
	B8-Guidance signs	0.25	0.34	0.18	0.18	0.05
	B9-Travel safety	0.23	0.34	0.31	0.10	0.02
A4-Empathy	B10-Accessibility service	0.30	0.36	0.22	0.08	0.04
	B11-Individual service	0.28	0.30	0.28	0.10	0.04
	B12-Passenger feedback	0.21	0.42	0.25	0.05	0.07
A5-Tangibility	B13-Cleanliness	0.15	0.41	0.26	0.14	0.04
	B14-Maintenance	0.22	0.35	0.25	0.15	0.03
	B15-Lighting system of station	0.19	0.39	0.29	0.08	0.05
	B16-Staff appearance	0.18	0.35	0.30	0.14	0.03
A6-Convenience	B17-Station location	0.15	0.35	0.26	0.20	0.05
	B18-Ticket service	0.20	0.29	0.23	0.25	0.03
	B19-Transferring	0.18	0.31	0.29	0.19	0.03

and

Table 7. Calculation results of passenger importance membership degree.

First-Level Indicator	Second-Level Indicator	Very Important	Important	General Important	Unimportant	Very Unimportant
A1-Reliability	B1-Punctuality	0.31	0.39	0.23	0.06	0.01
	B2-Operation schedule	0.32	0.41	0.21	0.04	0.03
	B3-Smooth driving	0.39	0.40	0.15	0.05	0.01
A2-Responsiveness	B4-Passenger needs	0.35	0.28	0.23	0.14	0.01
	B5-Route information	0.23	0.45	0.20	0.09	0.04
	B6-Adequate transport	0.28	0.38	0.17	0.16	0
A3-Assurance	B7-Staff attitude	0.22	0.30	0.28	0.15	0.05
	B8-Guidance signs	0.17	0.36	0.24	0.18	0.05
	B9-Travel safety	0.22	0.38	0.21	0.12	0.07
A4-Empathy	B10-Accessibility service	0.22	0.35	0.26	0.14	0.03
	B11-Individual service	0.17	0.39	0.29	0.14	0.01
	B12-Passenger feedback	0.26	0.27	0.30	0.15	0.02
A5-Tangibility	B13-Cleanliness	0.37	0.44	0.12	0.06	0.01
	B14-Maintenance	0.31	0.45	0.21	0.02	0.01
	B15-Lighting system of station	0.33	0.42	0.21	0.04	0.01
	B16-Staff appearance	0.29	0.48	0.19	0.01	0.03
A6-Convenience	B17-Station location	0.35	0.43	0.14	0.05	0.03
	B18-Ticket service	0.35	0.39	0.17	0.08	0
	B19-Transferring	0.35	0.40	0.17	0.08	0

.

Then, combining the indicator weight set W and the two membership degree matrices R, the comprehensive evaluation matrix B can be calculated by Equation (6).

Finally, the comprehensive evaluation matrix B is integrated with the scores of passenger satisfaction and importance corresponding to the five-level set (5, 4, 3, 2, 1); the fuzzy comprehensive evaluation scores of passenger satisfaction and importance for the Yibin ART service quality are obtained.

For instance, firstly, based on the indicator weights obtained from AHP in Table 5 and the satisfaction membership degree matrix in Table 6, the fuzzy evaluation vector b₁ for the first-level indicator A1-Reliability is constructed as follows:

$$\begin{array}{l}{b}_1=(0.3248,0.3470,0.3282)\left[\begin{array}{ccccc}0.23& 0.29& 0.27& 0.18& 0.03\\ 0.24& 0.35& 0.16& 0.20& 0.05\\ 0.17& 0.35& 0.29& 0.12& 0.06\end{array}\right]\\ \begin{array}{c}=(0.2138,0.3305,0.2384,0.1672,0.0468)\end{array}\end{array}$$

Then, the rest of the fuzzy evaluation vectors b₂–b₆ for the first-level indicators A2–A6 can also be calculated, and the comprehensive evaluation matrix B of the first-level indicators for the passenger satisfaction is as follows:

$$B=\left[\begin{array}{ccccc}0.2138& 0.3305& 0.2384& 0.1672& 0.0468\\ 0.2130& 0.3422& 0.1742& 0.1977& 0.0720\\ 0.2436& 0.3699& 0.2219& 0.1393& 0.0304\\ 0.2752& 0.3424& 0.2531& 0.0842& 0.0452\\ 0.1833& 0.3671& 0.2791& 0.1362& 0.0342\\ 0.1851& 0.3069& 0.2498& 0.2261& 0.0341\end{array}\right]$$

Next, based on the scores corresponding to the five satisfaction evaluation levels (5, 4, 3, 2, 1), the fuzzy comprehensive score of the first-level indicators for the satisfaction of Yibin ART service quality is calculated by Equation (7) as follows:

$$\begin{array}{l}S=B{V}^T=\left[\begin{array}{ccccc}0.2138& 0.3305& 0.2384& 0.1672& 0.0468\\ 0.2130& 0.3422& 0.1742& 0.1977& 0.0720\\ 0.2436& 0.3699& 0.2219& 0.1393& 0.0304\\ 0.2752& 0.3424& 0.2531& 0.0842& 0.0452\\ 0.1833& 0.3671& 0.2791& 0.1362& 0.0342\\ 0.1851& 0.3069& 0.2498& 0.2261& 0.0341\end{array}\right]\left(\begin{array}{l}5\\ 4\\ 3\\ 2\\ 1\end{array}\right)\\ \begin{array}{cc}& =\end{array}{\left(3.49,3.42,3.67,3.72,3.53,3.39\right)}^T\end{array}$$

Later, the comprehensive evaluation matrix B is integrated with the weight set W of the first-level indicators, and the fuzzy comprehensive weights Wc of the first-level indicators are calculated as follows:

$$\begin{array}{l}{W}_C=BW=\left[\begin{array}{ccccc}0.2138& 0.3305& 0.2384& 0.1672& 0.0468\\ 0.2130& 0.3422& 0.1742& 0.1977& 0.0720\\ 0.2436& 0.3699& 0.2219& 0.1393& 0.0304\\ 0.2752& 0.3424& 0.2531& 0.0842& 0.0452\\ 0.1833& 0.3671& 0.2791& 0.1362& 0.0342\\ 0.1851& 0.3069& 0.2498& 0.2261& 0.0341\end{array}\right](0.2913,0.2529,0.0666,0.06,0.1433,0.1859)\\ \begin{array}{cc}& =\end{array}(0.2095,0.3376,0.2299,0.1746,0.0478)\end{array}$$

Finally, based on the five satisfaction levels (5, 4, 3, 2, 1), the fuzzy comprehensive score for the overall satisfaction of Yibin ART service quality is calculated as S_C = 0.2095 × 5 + 0.3376 × 4 + 0.2299 × 3 + 0.1746 × 2 + 0.0478 × 1 = 3.48, which is between “general satisfied” and “satisfied”, and the overall satisfaction is generally upper middle level.

Similarly, combining the importance membership degree of indicators in Table 7 and the weight set in Table 5, the fuzzy comprehensive scores of the importance of the indicators in the first-level can be calculated as (4.01, 3.78, 3.50, 3.58, 4.03, 4.00). Further, the fuzzy comprehensive score of the overall importance for Yibin ART service quality is obtained as 3.89, which is above average and close to the evaluation level of “important”.

Therefore, based on the six-dimensional evaluation system of the extended SERVQUAL model, the overall satisfaction comprehensive score is 3.48, and the overall importance comprehensive score is 3.89. The service quality gap is 3.48 − 3.89 = −0.41, indicating that the passenger expectations for service quality are significantly higher than their actual perception, that is, the actual performance of the service quality does not meet the core needs of passengers. So, it is necessary to further analyze the service quality gaps between satisfaction and importance of all indicators and conduct a precise analysis.

4.3. Results Analysis

In order to further identify the specific weak points of the service indicators, the passenger satisfaction and importance evaluation scores of all indicators are calculated by using a membership degree matrix and the five-level evaluation set, and the final evaluation results of Yibin ART service quality are shown in

Table 8. Evaluation results of Yibin ART service quality.

First-Level Indicator	Weight	Service Quality			Second-Level Indicator	Weight	Service Quality
		Satisfaction	Importance	Gap			Satisfaction	Importance	Gap
A1-Reliability	0.2913	3.49	4.01	−0.52	B1-Punctuality	0.0946	3.51	3.93	−0.42
					B2-Operation schedule	0.1011	3.53	3.98	−0.45
					B3-Smooth driving	0.0956	3.42	4.11	−0.69
A2-Responsiveness	0.2529	3.42	3.78	−0.36	B4-Passenger needs	0.0674	3.40	3.85	−0.45
					B5-Route information	0.0248	3.38	3.77	−0.39
					B6-Adequate transport	0.1607	3.44	3.75	−0.31
A3-Assurance	0.0666	3.67	3.50	0.17	B7-Staff attitude	0.0332	3.72	3.49	0.23
					B8-Guidance signs	0.0119	3.56	3.42	0.14
					B9-Travel safety	0.0215	3.66	3.56	0.10
A4-Empathy	0.0600	3.72	3.58	0.14	B10-Accessibility service	0.0218	3.80	3.59	0.21
					B11-Individual service	0.0279	3.68	3.57	0.11
					B12-Passenger feedback	0.0103	3.65	3.60	0.05
A5-Tangibility	0.1433	3.53	4.03	−0.50	B13-Cleanliness	0.0314	3.49	4.10	−0.61
					B14-Maintenance	0.0317	3.58	4.03	−0.45
					B15-Lighting system of station	0.0142	3.59	4.05	−0.46
					B16-Staff appearance	0.0660	3.51	3.99	−0.48
A6-Convenience	0.1859	3.39	4.00	−0.61	B17-Station location	0.0384	3.38	4.02	−0.64
					B18-Ticket service	0.1053	3.38	3.98	−0.60
					B19-Transferring	0.0422	3.42	4.02	−0.60

below.

Table 8 provides an overall perspective of the passenger perception and expectation for the Yibin ART service quality, and the data in the table reflect the perception and expectation of each indicator in each dimension, as well as the gap score.

Firstly, based on the service quality gap values and indicator weights listed in Table 8, regarding the first-level indicators, it can be observed that, A1-Reliability has the highest weight of 0.2913 and a significant gap of −0.52, indicating that all aspects of indicator A1-Reliability need to be prioritized for improvement; A6-Convenience has a weight of 0.1859 and a significant gap of −0.61, indicating that all aspects of this dimension need to be actively improved; A5-Tangibility has a weight of 0.1433 and a gap of −0.50, indicating that the improvement is needed, with a focus on enhancing environmental cleanliness; A2-Responsiveness has a weight of 0.2529 and a gap of −0.36, indicating that dynamic scheduling capabilities urgently need to be strengthened; A3-Assurance has a weight of 0.0666 and a gap of 0.17, and A4-Empathy has a weight of 0.0600 and a gap of 0.14, with both weights being close and the service quality gaps being positive, indicating that these two dimensions are performing well overall. However, since the service gap of A4-Empathy (0.14) is smaller than that of A3-Assurance (0.17), the priority should be given to A4-Empathy.

Therefore, the priority for improvement in each dimension is as follows: A1-Reliability > A6-Convenience > A5-Tangibility > A2-Responsiveness > A4-Empathy > A3-Assurance. Among the secondary indicators corresponding to these primary indicators, the most urgent improvement indicator in A1-Reliability is B3-Smooth driving, with a weight of 0.0956 and a gap of −0.69, which refers to the smooth operation of the ART vehicle without abnormal bumps. The indicator in A6-Convenience that firstly needs improvement is B18-Ticket service with a weight of 0.1053 and a gap of −0.60, which refers to the ticket purchase should be easy so as to save time. In A5-Tangibility, the first improvement indicator is B13-Cleanliness with a weight of 0.0314 and a gap of −0.61, which means that the cleanliness of the station urgently needs improvement. In A2-Responsiveness, the indicator that needs to be improved first is B6-Adequate transport, with a weight of 0.1607 and a gap of −0.31, which means that there is a need for adequate vehicles to save passengers waiting time, especially during peak hours, on weekends, and public holidays.

Additionally, according to Table 8, the service quality gaps between passenger satisfaction and expectation are reflected in Figure 4, and the detailed analysis of key issues related to the second-level indicators in each dimension is as follows.

(1) A1-Reliability dimension results analysis

Firstly, the satisfaction rating for the B1-Punctuality is 3.51, which is lower than the importance level of 3.93, with a gap of −0.42. Secondly, the satisfaction rating for the stability of the B2-Operation schedule is 3.53, which is lower than the importance level of 3.98, with a gap of −0.45. This may be due to the interference from traffic lights at intersections and other vehicles, causing fluctuations in punctuality during certain periods and resulting in low passenger satisfaction. Additionally, the satisfaction rating for the operational stability of B3-Smooth driving is 3.42, which is lower than the importance level of 4.11, with a gap of −0.69. This may be due to the vehicle experiencing significant bumps when negotiating curves or uneven road surfaces, affecting passenger comfort.

(2) A2-Responsiveness dimension results analysis

Firstly, the satisfaction level of B4-Passenger needs is 3.40, which is less than the importance level of 3.85, with a gap of −0.45, reflecting the need to meet the passenger requirements. Secondly, the satisfaction rating of B5-Route information is 3.38, lower than the importance level of 3.77, with a gap of −0.39, indicating that the real-time information update of ART is slow, and passengers cannot grasp the real-time operation information of ART in a timely manner. In addition, the satisfaction rating for B6-Adequate transport is 3.44, lower than the importance level of 3.75, with a gap of −0.31. This may be due to the current fixed scheduling based solely on historical data, lacking dynamic adjustments under real-time passenger flow monitoring, and insufficient consideration for a flexible increase in vehicles during peak hours.

(3) A3-Assurance dimension results analysis

Firstly, the satisfaction rating of B7-Staff attitude is 3.72, higher than the importance level of 3.49, with a gap of +0.23, reflecting the advantages of ART service in humanistic services. Secondly, the satisfaction rating for B8-Guidance signs is 3.56, higher than the importance level of 3.42, with a gap of +0.14, indicating that the ART operation service process is simple and clear, making it easy for passengers to understand at a glance. In addition, the satisfaction rating of B9-Travel safety is 3.66, greater than the importance level of 3.56, with a gap of +0.10, indicating that the feeling of security of the passengers is high during the travel, and the safety measures meet the passengers’ expectations.

(4) A4-Empathy dimension results analysis

Firstly, in terms of providing accessible facilities in the B10-Accessibility service, the satisfaction rating is 3.80, higher than the importance rating of 3.59, with a gap of +0.21. Secondly, B11-Individual service has a satisfaction rating of 3.68 in terms of priority service measures for the elderly, pregnant women, and other groups, which is higher than the importance rating of 3.57, with a gap of +0.11. In addition, the satisfaction with the smooth feedback channel for B12-Passenger feedback is 3.65, which is close to the importance level of 3.60, with a gap of +0.05. This indicates that ART performs well in the dimension of empathy, not only meeting the requirements of passengers but even exceeding their expectations.

(5) A5-Tangibility dimension results analysis

Firstly, the satisfaction rating of B13-Cleanliness is 3.49, which is lower than the importance level of 4.10, with a gap of −0.61. This may be due to the delayed cleaning of garbage in the carriage during the morning rush hour, as well as the presence of cigarette butts, fallen leaves, and other debris in the gaps of the platforms. Secondly, the satisfaction rating of B14-Maintenance regarding the comfort of seats, air-conditioning, and other facilities is 3.58, which is lower than the importance level of 4.03, with a gap of −0.45, and it means that the inadequate maintenance of the equipment has resulted in an uncomfortable riding experience for passengers. In addition, the satisfaction rating of the B15-Lighting system of station is 3.59, lower than the importance level of 4.05, with a gap of −0.46. It is possible that some ground stations have insufficient night lighting brightness, which poses a safety hazard. Furthermore, the satisfaction rating of B16-Staff appearance is 3.51, which is lower than the importance level of 3.99, with a gap of −0.48. Maybe some staff dressed or behaved inappropriately, creating a negative impression on passengers.

(6) A6-Convenience dimension results analysis

Firstly, the satisfaction rating of the B17-Station location is 3.38, which is lower than the importance level of 4.02, with a gap of −0.64. This may be due to the station location not being reasonable, and the distance is too far. Secondly, the indicator B18-Ticket service has a satisfaction rating of 3.38, which is lower than the importance level of 3.98, with a gap of −0.60. Although there is a Yibin ART APP, the current ticket services mainly rely on a combination of manual ticket sales and self-service ticket machines, because there is always one artificial window, and the number of self-service ticket machines is relatively small, resulting in some elderly passengers queuing for long periods of time. So, the satisfaction with the ticket service is relatively low. In addition, the satisfaction rating of B19-Transferring is 3.42, lower than the importance level of 4.02, with a gap of −0.60. This may be due to the long walking distance required for transfer nodes and the long vehicle scheduling time, resulting in low transfer efficiency.

In summary, the indicators with a significant gap between expectation and perception of passengers include the B3-Smooth driving (with a gap of −0.69) regarding ART vehicles run smoothly without any abnormal bumps, B13-Cleanliness (with a gap of −0.61) regarding station and carriage environment being clean and no accumulation of debris, B17-Station location (with a gap of −0.64) regarding the rationality of station location, B18-Ticket service (with a gap of −0.60) regarding the tickets are convenient to purchase and the waiting time is short, and the B19-Transferring (with a gap of −0.60) regarding convenient transfer with other public transportation such as buses.

Therefore, the improvement suggestions are provided for these 5 indicators as follows:

(1) Regarding the indicator B3-Smooth driving, fault prediction and maintenance optimization can be used, including using quality analysis methods to analyze vehicle vibration sources, identifying failure modes of key components (such as shock absorbers and wheel bearings), developing a “Preventive Maintenance Plan”, shortening maintenance cycles, such as monthly inspection of suspension systems, quarterly calibration of wheel dynamic balance, and real-time warning of faults through the Internet of Things platform to achieve “predictive maintenance” and improve vehicle stability. In addition, vehicle-mounted vibration sensors are installed to monitor the acceleration of the vehicle in real time, such as setting a threshold to trigger an alarm when speeding. At the same time, in order to ensure the smooth operation of the ART vehicle, it is necessary to optimize the vehicle track and route. For example, virtual track flatness detection should be carried out on high bumpy sections such as bends and intersections, and elastic materials should be filled in time for out-of-tolerance sections. GPS trajectory analysis tools should be used to optimize driving strategies, such as decelerating through bumpy sections in advance and reducing shaking caused by sudden braking and acceleration.

(2) Regarding the indicator B13-Cleanliness, on-site 5S (i.e., Sort, Set, Shine, Standardize, and Sustain) management can be strengthened. Firstly, Sort, such as removing useless debris from blind spots, such as platform gaps and seat bottoms, and only retaining necessary cleaning tools (such as vacuum cleaners and brooms). Then, Set, such as setting up visual tool racks in the carriage cleaning cabinet and sorting tools according to “frequency of use” (such as placing garbage bags on the upper layer and tools on the lower layer). Next, Shine, such as developing a “Cleaning SOP (Standard Operating Procedure) Schedule” to clarify cleaning tasks at different times. Finally, Standardize and Sustain, which means developing standards and cultivating habits to achieve daily standard cleaning and improve staff literacy.

(3) For indicator B17-Station location, in the selection and design of the station location, various factors should be taken into account, such as the distance between the station and the main roads, the surrounding service facilities, and the capacity requirements of the station. It is recommended to choose the location close to densely populated areas such as commercial centers, residential areas, schools, etc., to facilitate passenger travel. In addition, the layout of the location should not be too far away.

(4) For the indicator B18-Ticket service, which is related to the convenience for ticket purchase, the ticket purchase process can be rebuilt, and full process automation is considered, such as adding the function of a face recognition gate, which supports “face swiping and entering the station with automatic fee deduction” (binding Alipay/WeChat). In addition, retaining one manual service window, only for cash ticket purchases and special needs (such as ticket refunds or loss reports), and guiding other passengers to self-service machines or mobile ticket purchases. Furthermore, considering the characteristics of elderly passengers, human–machine engineering optimization should be carried out for self-service machines, such as increasing the font size for visibility, and adding a voice prompt function, so that elderly passengers can synchronously play operation instructions when clicking on the screen.

(5) For the indicator B19-Transferring, transfer path simulation and optimization can be carried out, using simulation software to simulate passenger transfer behavior, identify bottleneck sections (such as congestion caused by insufficient width of a transfer channel), and optimize station layout. In addition, adding an intelligent guidance system to display real-time walking route (including distance and estimated time) through mobile AR function, and setting up illuminated guidance signs at key intersections (such as turns and staircases), especially the brightness should be sufficient at night, to ensure clear visibility.

In addition, the suggestions for improving other indicators are as follows:

(1) Suggestions for reliability-related indicators

Reliability-related indicators include the indicators B1-Punctuality, B2-Operation schedule, and B3-Smooth driving. In order to ensure the reliable operation of ART, practical and effective measures must be taken to deal with various situations that affect the normal, continuous operation of the ART and reduce the waiting time for ART. For example, the ART can be linked with the traffic police department to set up the “ART priority” signal mode at key intersections on ART rail lines to save ART vehicles waiting time. In addition, carrying out dynamic parking management, such as real-time counting of the number of passengers boarding and alighting through interior cameras, and intelligent adjustment of parking time, can save boarding and alighting time.

(2) Suggestions for responsiveness-related indicators

Responsiveness-related indicators include the indicators B4-Passenger needs, B5-Route information, and B6-Adequate transport. In order to ensure the needs of passengers and the stable operation of ART, practical and effective measures must be taken to respond to various requirements and emergencies. These measures include establishing a sound early warning system, improving emergency response plans, enhancing personnel training, and improving technical prevention capabilities. Through these efforts, the impact of emergencies can be minimized to the greatest extent possible, ensuring the safety of passengers and staff and ensuring the stable operation of the ART. For example, with the help of high-tech supported information management platforms and emergency response systems, smooth and unobstructed information transmission can be ensured in the event of various incidents, while ensuring timely and effective response measures. In addition, adopting an intelligent operation scheduling system to reduce faults or accidents caused by human negligence, and in response to large-scale cultural and sports events, concerts, and other situations, measures such as adding temporary vehicles and shortening vehicle intervals should be taken in a timely manner to maximize transportation capacity and effectively alleviate passenger flow pressure.

(3) Suggestions for assurance-related indicators

Assurance-related indicators include the indicators B7-Staff attitude, B8-Guidance signs, and B9-Travel safety. Firstly, for staff attitude, consider providing relevant training and coaching for staff to improve their work ability and confidence, which can help improve their work attitude. Then, for the guidance signs, it is better to place the signage in a prominent location, such as at the intersection of main passages, and distinguish it with signs of different colors or shapes. In addition to using text explanations, adding some simple and easily understood charts can help passengers understand at a glance. In addition, in areas with high pedestrian traffic, setting up electronic display screens to display real-time dynamic information of ART, such as the upcoming vehicle number and estimated arrival time, is beneficial. Further, to ensure the travel safety, it is necessary to strengthen the training of staff for emergency response, passenger service, and safety awareness, and some useful measures can be taken, including installing video surveillance equipment, installing security check equipment at the entrance and carriage entrance, and installing emergency assistance devices in the ART carriages and stations to help passengers receive timely assistance.

(4) Suggestions for empathy-related indicators

Empathy-related indicators include the indicators B10-Accessibility service, B11-Individual service, and B12-Passenger feedback. In order to improve empathy, the ART managers and staff should pay close attention to the transportation needs of key groups such as the elderly, young, sick, disabled, and pregnant, guide operating units to improve station accessibility facilities, optimize accessibility elevator guidance, actively provide one-stop travel guidance such as off-site organization, in station ticket purchasing, scanning at the gate, and boarding guidance, and fully guarantee the smooth transportation of key groups and improve service experience.

(5) Suggestions for tangibility-related indicators

Tangibility-related indicators include the indicators B13-Cleanliness, B14-Maintenance, B15-Lighting system of station, and B16-Staff appearance. In order to improve riding comfort, before the ART is put into operation, the focus should be on increasing the maintenance and cleaning of the air-conditioning system to ensure the normal operation of the ventilation and cooling systems, so as to provide passengers with a comfortable riding experience. In addition, the ART can take the measure of “strong cold and weak cold carriages” and add warning signs at the air outlets to allow passengers to choose suitable carriages according to their own needs. At the same time, the frequency of cleaning and sweeping at the entrances and exits, and platforms of stations along the rail transit line should be significantly increased, especially in key areas such as bathrooms and trash cans. For the staff’s appearance, it can be improved through training on standardized service etiquette and conducting regular inspections.

(6) Suggestions for convenience-related indicators

Convenience-related indicators include the indicators B17-Station location, B18-Ticket service, and B19-Transferring. Due to the significant gaps between the satisfaction and expectation of these indicators, the improvement suggestions for these indicators have been separately described in the previous text, mainly including optimizing and adjusting the route operation plan, installing the advanced intelligent systems to improve ticket purchasing efficiency, comprehensively optimizing the orientation of entrances and exits, and clearly labeling surrounding information such as landmark buildings, government centers, office buildings, and shopping malls to help passengers plan their travel and transfer routes more clearly.

By implementing these suggestions and measures, the goal can be achieved to create a more efficient, convenient, environmentally friendly, and comfortable ART travel experience for passengers.

4.4. Discussion

This study aims to investigate the service quality perception and expectation of ART users in Yibin City. Research has found significant differences between participants’ perceptions and expectations for ART service quality, leading to dissatisfaction with the services they received, especially those deemed unreliable and inconvenient. According to the survey results and suggestions, we can speculate that the research findings can demonstrate and promote the potential improvements in the operation of ART.

Firstly, by combining the AHP with the FCE method, the evaluation is conducted, and the results are obtained. The study has shown that the core of the ART service quality is the service reliability, with a weight of 0.2913, and the responsiveness, with a weight of 0.2529. In addition, the newly added attribute A6-Convenience, with a weight of 0.1859, indicates that passengers are concerned and have a demand for convenience services, which is also consistent with the investigation results.

Further, the investigation results show that the overall satisfaction score of Yibin ART service quality is 3.48, and the overall importance score reaches 3.89, with a gap of −0.41, indicating that the actual service performance does not fully meet the core needs of passengers. Among them, the differences in A1-Reliability, A6-Convenience, and A5-Tangibility dimensions are particularly significant, such as the gap between the passenger satisfaction and importance for indicator B3-Smooth driving is −0.69, the gap of B17-Station location is −0.64, the gaps of B18-Ticket service and B19-Transferring are both −0.60, and the gap of B13-Cleanliness is −0.61. All these issues of Yibin ART service quality are exposing the shortcomings of Yibin ART in operational efficiency and convenience, and the conclusion is consistent with the problems and needs reflected in the references [7,9,46,47].

Finally, through evaluation, the critical service improvement needs are identified, and the targeted improvement measures have been proposed for these key service indicators, which can guide the future ART service modifications. Doing this will not only serve the interests of the current users but also attract potential users, resulting in increased ART use, which will have a positive impact on the traffic congestion and related air pollution in the city.

5. Conclusions and Future Research

5.1. Conclusions

The ART is a cross-border integration product, which has significant characteristics such as high-speed rail technology, subway management, and flexible public bus transportation. As a new medium-capacity urban rail transit system and a relatively new transit mode, applying the integrated SERVQUAL–AHP–FCE framework to ART has important theoretical and practical significance for improving the quality of ART services.

This study is based on the SERVQUAL model, combined with the AHP and FCE, to systematically evaluate and analyze the service quality of ART. Firstly, based on the SERVQUAL model and considering the operational characteristics of Yibin ART, a new attribute “convenience” is added to extend the SERVQUAL model and form a service quality evaluation indicator system that includes 6 dimensions of reliability, responsiveness, assurance, empathy, tangibility, and convenience, as well as 19 secondary indicators. Then, a comprehensive evaluation method integrated AHP with FCE for assessing the Yibin ART service quality is described, and during the evaluation process, the indicator weights are determined by using AHP, and the FCE method is used to evaluate and quantify the service quality gap between passenger satisfaction and expectation. Finally, through the analysis of the evaluation results, the important indicators that have a significant impact on Yibin ART service quality are identified, and the indicators with significant differences in service quality between passenger satisfaction and expectation have been found. Simultaneously, some targeted suggestions and strategies for improving the Yibin ART service quality are proposed. The main conclusions of the research can be summarized into four aspects.

(1) Construction and verification of the theoretical framework

Firstly, considering the passenger needs, an extended SERVQUAL model for evaluating Yibin ART service quality is constructed with reliability, responsiveness, assurance, empathy, tangibility, and convenience. In addition, for the established six-dimensional evaluation indicator system of Yibin ART service quality, the effectiveness of the evaluation model is validated with the questionnaire survey data by statistical analysis.

(2) Quantitative analysis of service quality gap

The FCE method is used to comprehensively evaluate the Yibin ART service quality, which can transform qualitative evaluation into quantitative evaluation based on the membership theory of fuzzy mathematics. Therefore, the quantitative service gap is obtained. For example, the overall satisfaction score of Yibin ART service quality is 3.48, and the expectation score is 3.89. The satisfaction score is lower than the expectation score, and the service gap is −0.41, indicating that the actual service performance does not fully meet the expectations of passengers. Further, by quantitatively calculating the service gap, the priority for improving the service indicators can be accurately determined, such as the indicators that urgently need improvement are B3-Smooth driving (with a gap of −0.69), B13-Cleanliness (with a gap of −0.61), B17-Station location (with a gap of −0.64), B18-Ticket service and B19-Transferring (both with a gap of −0.60). So, the quantitative analysis of the service quality gap can provide decision-makers with a more objective reference basis.

(3) Identification of key issues and proposal of strategies

The study proposes the specific improvement measures for indicators with significant gaps, such as optimizing vehicle maintenance through onboard sensors, implementing 5S management to improve station environment, optimizing the selection and design of station location, adding intelligent ticketing equipment to shorten waiting time, and adding intelligent guidance system to improve the transferring efficiency, so as to narrow the gap between service supply and passenger expectation.

(4) Dual contributions of theory and practice

The main contributions of the study include the dual contributions of theory and practice. At the theoretical level, the study has expanded the application of the SERVQUAL model in the field of ART, providing a reference framework for service quality evaluation of similar new transportation systems. At the practical level, the evaluation method and proposed optimization strategies provide a scientific basis for the operation and management of Yibin ART, and also accumulate experience for the development of ART in other cities, helping them to promote the sustainable development of urban public transportation, ultimately promoting low-carbon and sustainable development of the entire socio-economic system.

5.2. Limitations and Future Research

Although there have been some achievements in the evaluation and quantitative analysis of ART service quality in Yibin City, there are still some challenges that can be further studied.

Firstly, the sample of this study mainly focuses on the local passengers in Yibin, and considers universality with a sample size of 110, without sufficient coverage of sub-groups such as tourists, commuters, and elderly people. There are differences in the demand and perception of ART services among different groups. For example, tourists may be more concerned about route guidance and transfer convenience, while the elderly have higher requirements for ticket purchasing convenience and barrier-free facilities. The singularity of the sample may result in evaluation results that cannot fully reflect diverse needs. So, the future research will expand the sample and add segmented groups such as tourists, commuters, and people with disabilities, design differentiated questionnaires, and analyze the differences in needs among different groups. For example, increasing the “convenience of connecting tourist attractions” indicator and strengthening the “experience of using accessible facilities” survey for people with disabilities. Additionally, the analysis of sample reliability needs to be strengthened. Currently, the survey only adopts a simple random sampling method, selecting a sample size that meets statistical analysis requirements. In future research, the more reasonable probability sampling methods, such as stratified random sampling, will be considered to further improve sample representativeness. Simultaneously, combining confidence level and error range to determine the sample size, avoiding situations where the sample size is too small and may not reflect the overall characteristics.

Secondly, the study adopts a static weight system and does not dynamically adjust the weights of indicators based on real-time passenger flow data (such as morning and evening peak hours and holiday passenger flow fluctuations). For example, during peak hours, passengers may have a significantly higher demand for “stability of vehicle intervals” and “responsiveness,” while during off-peak hours, they are more concerned with “comfort of the riding environment”. In addition, the model does not consider all the factors that affect service quality and analyze the relationship with service quality. So, the future research will use the deep learning models to analyze the historical service data and predict passenger satisfaction trends, and employ an association rule mining algorithm to identify key factor combinations that affect service quality satisfaction. So, future research can conduct in-depth research on passenger flow dynamics and travel characteristics, carefully grasp the travel needs of passengers, continuously optimize operational service plans, and strive to improve the service level of ART.

Finally, the study only surveys the people who have used the ART, but the potential passengers who do not use the ART are not included. Future research will collect the reasons why they do not use ART, so as to improve the service limitations and increase the potential passengers’ use of ART. Additionally, the study only focuses on routine operational scenes and does not address the impact of unexpected events (such as equipment failures, extreme weather) on service quality. For example, the punctuality rate, emergency response capability, and passenger satisfaction of the ART may fluctuate significantly in rainstorm weather, but the existing evaluation system does not carry out a special assessment in such a scenario, lacking analysis on service resilience. So, in the future, evaluation service scenes will be expanded and pay more attention to the emergency service capability, so as to promote the stable and sustainable development of ART.