Identification of the regional and economic contexts of sustainable urban logistics policies

: Urban logistics policies have become instrumental in achieving sustainable transport systems. Developing and emerging countries still lag far behind in the implementation of such policies when compared with developed countries. This exposure gap provides an opportunity for policy transfer, but this is a complex process requiring knowledge of many contextual factors and involving multiple steps. A good understanding of those contextual factors of measures by cities may be critical for a successful transfer. Our study aimed to identify the different contexts of urban logistics measures or policies worldwide and to assess their significance for policy transferability. In this study, urban logistics measures discussed in the literature were retrieved with a systematic literature review method and then the contexts were recorded, distinguishing between economic development levels and geographical regions. The analysis revealed that the economic level and geographical location of cities both have a strong association with the type of measure implemented. Barriers and drivers were identified by assessing policy transfer between developed and developing countries. Institutional and physical barriers appeared to be highly pertinent for a range of measures, while drivers or facilitators were identified from specific problems in developing countries and the respective measures in developed countries. Thus, the analysis of contextual factors can provide a first response to the key challenges and opportunities of sustainable urban logistics policies transfer to developing countries. can be simpliﬁed by applying a systematic literature review (SLR) method, which uses well-deﬁned and rigorous criteria to identify existing studies and select contributions to further analyze and synthesize [30,31]. A step-by-step SLR approach was used here to retrieve relevant articles. and the speciﬁc “measures” or “solutions” from the database search were from search results. inclusion / exclusion criteria were applied through and content analysis to with the focus on sustainable urban logistics measures.


Introduction
Today, more than half of the world's population resides in urban areas [1]. The increase in urban population and continued economic growth create a necessity for advanced urban logistics (UL) schemes [2]. These schemes help to sustain and maintain the urban way of life but also impose a range of external impacts in terms of congestion, air pollution, accidents, and noise [3]. Several trends drive the need for urban logistics, such as the rise of e-commerce and the sharing economy, the desire for speed in supply chains, and increased attention to sustainability [4]. Freight transport in urban areas is inherently complex, due to its multiple stakeholders, often with conflicting stakes [5], intricate routing patterns, and diverse goods types [6]. Urban logistics as a discipline specializes in coping with the sustainability problems encountered in urban freight transport [7]. The adoption of measures or policies that promote sustainability is thus essential [8].
Urban logistics challenges differ between cities in different regions of the world. For instance, cities in developed countries exhibit rapid changes, with lower store inventories and just-in-time (JIT) supplies to businesses, a significant increase in the type and variety of products on the market with the The remainder of the paper is structured as follows. Section 2 describes the research methodology used in the analysis, Section 3 presents the results obtained, and Section 4 discusses the main findings. Section 5 provides some concluding remarks.

Materials and Methods
The process of policy transfer involves a range of variables and multiple steps. The most common extensive framework, by Dolowitz and Marsh [24], comprises key policy transfer components and seven questions: (1) What is transferred? (2) Why do actors engage in policy transfer? (3) Who are the key actors involved in the policy transfer process? (4) From where are lessons drawn? (5) What are the different degrees of transfer? (6) What restricts or facilitates the policy transfer process? Finally, (7) how is the process of policy transfer related to policy "success" or policy "failure"? Addressing these questions provides a clear and complete picture of the feasibility of policy transfer. Several studies have applied the framework in the transport domain [19,21,22,25,26].
Macario and Marques [18] suggest a 10-step logical framework for the transfer of urban mobility measures which is capable of addressing the seven questions ( Figure 1). Their framework sets the sequence and interrelations between various steps, to identify the potential for a successful transfer.
The work in this study was organized into four stages (I-IV), to cover the contextual variability in sustainable urban logistics policies around the world, and linked to five steps in the 10-step framework, as illustrated in Figure 1. Knowledge of the contexts of urban logistics measures can be useful input to cities when establishing the main problem/s (Step 1), looking for similar contexts (Step 4), and deciding on the practices to be adopted (Step 5). Identification of barriers and drivers specifies the particular assessment points of the measure(s) (Step 8) and provides insights when adjustments are required (Step 9).

Stage I: Retrieve Relevant Urban Logistics Measures from the Literature
The urban logistics literature is currently showing a steep increase in terms of both the number of studies performed and the diversity of discussion topics [5,[27][28][29]. The challenge of retrieving representative results from existing studies can be simplified by applying a systematic literature review (SLR) method, which uses well-defined and rigorous criteria to identify existing studies and select contributions to further analyze and synthesize [30,31]. A step-by-step SLR approach was used here to retrieve relevant articles.
Keywords with the general term "urban logistics" (or other synonyms such as city logistics, urban delivery, urban freight transport, urban goods transport, urban goods distribution, urban goods delivery, last-mile logistics, and last-mile delivery) and the specific terms "measures" or "solutions" were combined with Boolean logic to formulate the search string. Two databases were searched, Scopus and Web of Science Core Collection, with the search restricted to peer-reviewed journal articles in the English language published after the year 2000. Relevant articles were retrieved from the database search and the duplicates of similar articles were removed from the search results. Then, inclusion/ exclusion criteria were applied through abstract and content analysis to filter articles with the focus on sustainable urban logistics measures.

Stage II: Filter the Measures into Regional and Economic Development Categories
The retrieved and selected urban logistics measures were categorized based on geographical/ regional location and economic development class of the source. The source of urban logistics measures in the papers reviewed was set as the country of the case study (if applicable) in the first instance, and then affiliation of the first author. The regional categorization followed the six continents (Africa, Asia, Oceania (including Australia), Europe, North America, and South/Latin America, with Antarctica excluded since it has no permanent human habitation).  The economic development categorization was based on a United Nations [32] report that divides countries into three broad categories: developed economies, economies in transition, and developing economies. Here, emerging market economies were considered as the fourth category in the discussion of the results. Emerging market economies are striving to become advanced and are generally on a more economically disciplined track. However, no universal consensus has yet been reached on exactly which countries qualify as emerging market economies. Different financial institutions have made up lists, including the International Monetary Fund (IMF), Dow Jones and Russell, and Morgan Stanley Capital International (MSCI)'s Emerging Market Index [33]. The MSCI list is the most comprehensive, comprising 26 countries, all of which were selected for this study (Argentina, Brazil, Chile, China, Colombia, Czech Republic, Egypt, Greece, Hungary, India, Indonesia, Korea, Malaysia, Mexico, Pakistan, Peru, Philippines, Poland, Qatar, Russia, Saudi Arabia, South Africa, Taiwan, Thailand, Turkey, and United Arab Emirates). Therefore, four economic development classes were recognized in this study: developed economies, emerging economies, economies in transition, and developing economies. Countries such as the Czech Republic, Greece, Hungary, and Poland, which can be categorized as both developed and emerging economies, were considered to be developed economies in this study.

Stage III: Establish the Contexts of the Measures with the Categories
The next stage involved two main tasks, identifying the main topics addressed and then distinguishing the contexts of the measures. Topic identification followed the method/approach used by Lagorio et al. [34], setting the dominant theme of measures or policies in the papers with the main emphasis on their application areas. Topics refer to the type of solution(s) being discussed in the article as recognizable by the title, keywords, or abstract. This approach is flexible in allowing multiple topics per article, which helps to attain broad visualization in establishing the contexts.
The inclusion of more factors in the analysis of contexts improves the credibility of policy transfer [16]. The task of distinguishing contexts was carried out using the topics identified as a basis and outlining three components of the papers reviewed:

•
The main objective/research problem stated; • The methodology used to attain the objective or solve the problem; • Related considerations connecting the above two (objective-method combinations).
Each topic corresponded to at least one context. The contexts indicated the level of uptake of measures through their variability for each category.

Stage IV: Identify Barriers and Drivers in the Transfer of Measures between Contexts
Identification of barriers and drivers can facilitate the adoption of policies or measures between different economic development and geographical categories which exhibit different levels of exposure to urban logistics. Barriers are factors that undermine the success of policy transfer in the recipient city, while drivers are enablers or motivating factors/conditions or facilitators in the policy transfer. In this study, we used a more general approach both to generalize the contexts in the economic development categories, and then identify potential barriers and drivers in policy transfer across those categories.
TURBLOG [12] identified seven categories of barriers: financial, physical, technological, cultural, political, legal, and security. In this study, we considered four of these barrier types: financial, physical, technological, and cultural, for analysis of the respective topics. The other three categories (political, legal, security) are highly subject to local conditions of the city, or in general of the country.
The financial barrier accounts for the high cost of a measure in the recipient city. The natural or built aspects of the recipient city that make transfer unsuitable represent a physical barrier. Technological barriers include the unavailability of technological elements or deviation from the existing system. Cultural barriers involve the requirement to depart from the traditional culture operating in the recipient city through the measures [21]. There may also be institutional barriers, generated by the need to change the institutional framework and enhance its strength to implement and enforce a specific measure. The corresponding drivers are the opportunities that can be achieved through the transfer of measures to the recipient city.
In the analysis to identify barriers and drivers to transfer measures between developed and developing economies, we followed a more general approach. The basis for the analysis was the contexts of measures from developed economies and relevant sets of problems related to urban logistics in developing economies. However, due to the lack of literature describing pertinent problems in the case of developing economies, we had to conduct additional reviews of other published and gray literature. We rated the five barrier categories into three levels that indicated the necessity of transfer as: highly relevant, relevant, and irrelevant. Barriers rated highly relevant could strongly determine the success of policy transfer and possibly impose the highest negative impact on the transfer. Barriers rated irrelevant were inapplicable to the measure under discussion. Drivers were derived as potential opportunities or suitable conditions in the recipient city that could be obtained through the transfer of policy measures.

Urban Logistics Measures and Topics
The database searches in Scopus and Web of Science Core yielded a total of 387 articles after the removal of duplicates. Application of inclusion/exclusion criteria to filter for articles with the focus on urban logistics measures resulted in 325 articles, which were further used for classification and context analysis.
Urban logistics measures covered by the articles were classified into different geographical and economic development levels based on the case study used in the article (if any) or the country of affiliation of the first author (Table 1). This classification of measures revealed the exposure level or the distribution of measures across the geographical regions and economic development classes. The majority of the measures were in Europe, in developed economies. The last row in Table 1 indicates the exposure level to urban logistics measures across the economic development classes. Topic identification involved outlining the dominant theme in the measures or policies mentioned in articles and their application area (see Section 2). A single article might comprise more than one topic; our analysis allowed up to four different topics to be identified. These classifications resulted in a total of 19 different topics, covered 343 times. A full list of the topics, their detailed description, and the frequency of occurance is presented in Table A1 (in the Appendix A to this paper). The top five topics in terms of the total number of relevant articles were: carrier operations optimization, stakeholder participation, solution performance, sustainability, and urban consolidation centers (UCCs). As with published articles on urban logistics measures, the topics of discussion also varied greatly across different regions and economic development classes.

Urban Logistics Topics in Different Economic Development Classes
The topics covered revealed significant variations between the different economic development classes, with 260 topics from developed economies, four topics from economies in transition, 68 topics from emerging economies, and 11 topics from developing economies.
The economic development class context was analyzed in terms of the distribution of topics and the detailed focus on individual topics. The analysis indicated that 80% of the topics in developed economies were from Europe ( Figure 2), and the central topics were carrier operations optimization, stakeholder participation, and solutions performance. In emerging economies, around 66% of the topics were contributed by Asia (Figure 3), and the main topics were carrier operations optimization, stakeholder participation, and sustainability. The main topics found in economies in transition were stakeholder participation and sustainability. The topics in developing economies were crowd-shipping, urban consolidation centers (UCC)/distribution centers (UDC), e-commerce delivery, loading/unloading areas, alternative modes, and parcel lockers.
The context of the topics was established from the measures' objective(s) and/or method(s), or combinations, used to solve the prevailing problems. These contexts can be categorized under the same topic while being found in different economic development classes. The major highlights of urban logistics contexts in the different economic development classes are presented in Table 2.

Urban Logistics Topics across Geographical Regions
Analysis of regional context by comparing different topics across geographical regions (continents, i.e., Africa, Asia, Australia (representing Oceania), Europe, North America, and Latin America) revealed that their contribution varied significantly (Table 1). The analysis involved checking the topics of discussion across the geographical regions, and also within the same economic development classes for more clarity (Supplementary Table S2). Europe contributed the most topics, while Africa contributed the least.

Regional Contexts in Developed Economies
Europe contributed to all 19 topics identified. North America contributed to the topics solution performance, stakeholder participation, pick-up points, off-hour delivery (OHD), cargo bikes, carrier operations optimization, traffic congestion management related to UL, and crowd-shipping. Asia contributed to the topics of road pricing, solution performance, e-commerce delivery, and carrier operations optimization. Australia contributed to the topics solution performance, stakeholder participation, alternative modes, eco-friendly vehicles, UCCs, and traffic congestion management related to UL. The main findings were: (1) Traffic congestion management related to UL: The context in all regions with developed economies was on solving congestion to improve delivery performance. (2) Carrier operations optimization: The objective was generally to optimize routing and other operations by the carrier. This was similar for all regions but differed in the problem/s tackled. The contexts in Europe involved single-and multi-echelon routing, electric vehicles, reducing emissions, truck-based drones, and intelligent transport systems. In North America, the contexts were time windows, network optimization, deployment of drones and eco-friendly vans, and carrier collaboration. The context in Asia was time windows. (3) Stakeholder participation: The contexts in Europe involved stakeholder role, participation/preference, policy evaluation, and the multi-actor multi-criteria assessment (MAMCA) method for structuring stakeholder consultations. The context in both North America and Australia was collaborative decision support approaches.
Europe and Asia also contributed to the topics of e-commerce delivery and road pricing. The contexts in Europe on e-commerce delivery topics were e-retail experiences of customers, home delivery, and the engagement of local authorities in facilitation, whereas the contexts in Asia were policies to address home delivery issues. The contexts for road pricing topics in Europe were acceptance and equity, whereas the Asia context was the effects of pricing on actors. Europe and Australia contributed to the topics of alternative modes and UCCs. The context in Europe for the alternative modes topic was on the use of urban intermodal, passenger rail network, and urban waterways, whereas the Australian context was urban intermodal for a port city. European contexts on UCC topics were perceptions of users and related benefits, policies and governance mechanisms, and integration of electric vehicles, whereas Australia focused on connecting UCCs with eco-friendly vehicles.
Both North America and Europe contributed to the topics cargo bikes, off-hours delivery (OHD), and crowd-shipping. The European context was evaluating the use of electric cargo bikes from policy perspectives, whereas the North American context was the overall performance and economic viability. The European contexts on OHD topics were addressing stakeholder acceptability and evaluating policy gaps, whereas North America emphasized impact analysis. The context for crowd-shipping topics in Europe was existing mass transit systems, whereas the context in North America was assigning individual customers to the crowd-shipper.
In addition to the topics covered in common with the other regions, unique topics discussed in Europe were information and communications technology/intelligent transport systems (ICT/ITS), loading/unloading area, parcel lockers, policy making, sustainability, and limited traffic zones (LTZ).
Sustainability 2020, 12, x; doi: FOR PEER REVIEW www.mdpi.com/journal/sustainability mechanisms, and integration of electric vehicles, whereas Australia focused on connecting UCCs with eco-friendly vehicles. Both North America and Europe contributed to the topics cargo bikes, off-hours delivery (OHD), and crowd-shipping. The European context was evaluating the use of electric cargo bikes from policy perspectives, whereas the North American context was the overall performance and economic viability. The European contexts on OHD topics were addressing stakeholder acceptability and evaluating policy gaps, whereas North America emphasized impact analysis. The context for crowdshipping topics in Europe was existing mass transit systems, whereas the context in North America was assigning individual customers to the crowd-shipper.
In addition to the topics covered in common with the other regions, unique topics discussed in Europe were information and communications technology/intelligent transport systems (ICT/ITS), loading/unloading area, parcel lockers, policy making, sustainability, and limited traffic zones (LTZ).

Regional Contexts in Emerging Economies
All the contributions of emerging economies were from Asia and Latin America. The topics crowd-shipping, UCCs, and policy making were unique to Asia, and cargo bikes to South America ( Figure  3). The Asian context for e-commerce delivery topics was order fulfillment, whereas in South America the focus was emissions analysis on the use of local delivery points. The context for stakeholder participation in both Asia and South America was perceptions of the actors, while that for solution performance was adopting or transferring urban logistics solutions between cities. Similarly, emission footprint and sustainability plans were sustainability topics.

Regional Contexts in Emerging Economies
All the contributions of emerging economies were from Asia and Latin America. The topics crowd-shipping, UCCs, and policy making were unique to Asia, and cargo bikes to South America (Figure 3). The Asian context for e-commerce delivery topics was order fulfillment, whereas in South America the focus was emissions analysis on the use of local delivery points. The context for stakeholder participation in both Asia and South America was perceptions of the actors, while that for solution performance was adopting or transferring urban logistics solutions between cities. Similarly, emission footprint and sustainability plans were sustainability topics.

Regional Context in Other Economic Development Classes
The European countries in the economies in transition class only had two topics of discussion, stakeholder participation and sustainability. In developing economies, only a single topic on alternate modes was contributed by Africa, while the rest of the topics (crowd-shipping, UCCs, e-commerce delivery, parking areas, and parcel lockers) were covered by articles from Asia.

Regional Context in Other Economic Development Classes
The European countries in the economies in transition class only had two topics of discussion, stakeholder participation and sustainability. In developing economies, only a single topic on alternate modes was contributed by Africa, while the rest of the topics (crowd-shipping, UCCs, e-commerce delivery, parking areas, and parcel lockers) were covered by articles from Asia.

Barriers and Drivers of Transfer between Developed and Developing Countries
Transfer of measures is a multi-step and multi-variable process (Figure 1), where matching contexts is the core activity. In this study, we incorporated different contexts into transferability analysis to increase the reliability and success of the process. Among the range of relevant contexts, those in the source city and recipient city are of dominant importance. However, our analysis based on the identified topics revealed a variety of contexts for measures in different regions, even under the same topic. This variability made it necessary to identify the barriers (potential undermining factors) and the drivers (potential motivating factors) of success in policy transfer.
The identification of barriers and drivers was based on general problems in developing economies and the contexts of measures implemented in developed economies, as described in Section 3.1 and 3.2. However, the lack of published literature on urban logistics in developing economies meant that we had to seek additional information from other literature sources, such as gray literature. The review results were summarized into relevant general urban challenges and particular urban logistics problems (see Table A2 in the Appendix A). The problems ranged from institutional roles and set-ups to specific urban logistics operations. The general problems for the city development group were urban development and land use, institutional set-up and role, policy and planning, and general traffic systems, outlining the city traffic conditions. The problems related to specific urban logistics were grouped into three categories: freight infrastructures, urban logistics operations, and environmental impacts. These challenges and realities were directly used in the analysis of barriers and drivers.
Five barrier types were identified considering their suitability to the analysis in this study (as described in Section 2, Stage IV). Those barriers were rated for relevance in the transferability of the topic under consideration between the two contexts. The rating of relevance has three scale levels as highly relevant, relevant, and irrelevant. The ratings are given based on the contexts where the urban logistics topics measures were implemented at the source city in the developed economies and the existing challenges of cities in developing economies. Moreover, drivers or facilitators are the potential opportunities or suitable conditions in the recipient city that could be obtained or uncovered through the transfer of policy measures. The barriers, with their corresponding relevance scaling, and the drivers identified are shown in Table 3.

Discussion
Policy transfer concepts can be important to both practitioners and researchers in the urban logistics domain, but there is a limited amount of literature on this topic. Related studies, e.g., in the transport domain, call for more research on the role of contexts and the overall transfer process [35,36]. Establishing the context for successful urban logistics measures is considered a core task and vital step in the process. Many relevant contexts need to be considered in policy transfer, with the accuracy of the process reported to increase with an increasing number of contextual factors accounted for in the analysis [16]. In the present analysis, we examined the contexts of urban logistics measures for economies at different stages of development and in different geographical regions (continents). Two types of contexts were considered in all cases: the context in the source city and contextual factors in the recipient city. Establishing a broad context of measures in the source city can expedite policy transfer.
The results provide responses to three of the seven questions posed by Dolowitz and Marsh [24], namely: "What is transferred?"; "From where are the lessons drawn?"; and "What restricts or facilitates the policy transfer process?". The "bottom-up" approach of Timms [25] was used to consider policy transfer from a "city perspective". The results also contributed to five major steps in the 10-step logical framework devised by Macario and Marques [18] (Figure 1). The broad contexts of urban logistics measures contributed to the search for similar contexts (Step 4), selecting urban contexts in the source city (Step 5), and identifying potential measures for their transferability (Step 6). The identified barriers and drivers in the analysis to transfer measures contributed to ex ante assessment (Step 8) and to identifying the need for adjustment of measures (Step 9) to improve transferability between developed and developing economies.
The contexts were found to be highly different across geographical regions, even for economies within the same development class. These differences can be explained by taking the case of Europe and the USA, where the contexts for sustainable urban logistics measures are quite different [37]. For instance, the European context for the OHD topic was stakeholder acceptability [38,39], whereas the USA context was the overall impact of OHD programs [40], which were rated as having medium applicability to the USA context [37].
European articles also covered unique topics of discussion, such as ICT/ITS, LTZ, loading/unloading area, parcel lockers, and policy making. The applicability rating of some of these topics to the case of the USA was highly variable [37]. LTZ was rated highly effective, but has low applicability to the USA context, while ICT/ITS was rated medium for both effectiveness and applicability.
In addition to the contributions from different economic development classes, the regional contributions on topics varied widely. Europe contributed most to the relevant literature and Africa contributed the least, confirming findings by Kin et al. [13] that the Middle East and Africa are underexposed to urban logistics measures. Policy transfer has a crucial role to play in bridging this exposure gap. The policy transfer process involves the identification of potential barriers and drivers. Five general barriers were considered in this study (institutional, financial, physical, technological, and cultural), and identification of barriers for each topic was based on prevailing problems in developing countries (Table A2). The drivers considered were the general opportunities or gains in the recipient city.
Among the barrier types considered, institutional barriers were rated highly relevant for most of the topics considered, as a strong institutional set-up is required for the transfer process but a weak institutional framework generally prevails in recipient countries. The next major barrier was physical, reflecting a lack of adequate infrastructure. The high relevance of these barriers is in agreement with Pojani [41], who identified institutional, financial, and physical barriers to the transfer of sustainable urban transport practices to Southeast Asian cities such as Jakarta, Manila, Kuala Lumpur, and Bangkok. A range of general drivers was also identified, most providing opportunities to increase the desirability of policy transfer from the specific case to cities in developing economies (Table 2).
Further comparisons between our findings and those in similar studies with different contexts revealed the broad reality and specific conditions. For example, Maria and Marcus [20] studied the transfer of road pricing policy from other European cities to Valletta, Malta, and identified the unique geography of the city as a barrier, while international events, local conditions, and a political champion driving change were seen as specific drivers. Macario and Marques [18] found high sensitivity to the barriers of political commitment and strategy, legislation and regulative institutions, information and public relations, and technology. Besides, our study identified financial and physical barriers as highly relevant, in addition to institutional barriers, whereas the need to increase available investments was considered a general driver.
In the case of UCCs, Nordtømme et al. [42] identified financial concerns and stakeholder acceptability as the main barriers to UCC implementation in Norway (Europe). We also found physical and financial barriers to be highly important for this topic, along with institutional barriers. In general, policy transfer faces a range of barriers when dealing with different topics and dimensions. Rye et al. [26] claim that language and planning traditions are even crucial barriers to transport policy transfer between European cities. TURBLOG (2011) recommended two approaches to overcome barriers, one by adapting the transferability of measures to remove or reduce the aspect weakened by the barriers, and the other by combining one or more measures (policy package) to counteract the barriers.
Drivers (enablers or facilitators) of the transfer of sustainable urban logistics measures to cities in developing economies were established from the particular problems and respective topics in developed economies (Table 2). In some cases, factors that acted as a barrier to one measure were a driver of other measures. A good example was the presence of many small traditional retail stores or nano-stores in cities in developing economies, which complicates delivery and was identified as a barrier to carrier optimization of operations, but nano-stores can also be a driver in the implementation of pick-up points by acting as pick-up stations. However, the identification of specific drivers required a detailed focus on the particular contextual factors of the recipient city. For example, Timms [21] identified a history of cooperation between local authority and stakeholders as enablers for the transfer of regulative measures to the city of Cariacica, Brazil.
Overall, the findings of this study contribute to portray worldwide contexts of sustainable urban logistics measures to analyze the transferability of policy experiences between different contexts. Theoretically, this study implies the strong association of the economic development level and geographical location of a city with the context in which measures were implemented. Besides, the categorization of urban logistics measures indicated which topics were given more attention and at which geographic locations.
Furthermore, this study has a practical implication to authorities considering policy transfer in urban logistics. The contextual factors analysis can provide a first response to the key challenges and opportunities for the transfer of sustainable urban logistics policies, especially to the ones in developing countries. Moreover, the establishments of barriers and drivers specific to urban logistics measures or policies also enhance the transferability process and help to focus only on the important and relevant factors.
This study has limitations in covering all available urban logistics solutions worldwide due to two reasons, namely limited literature published in scientific journals generally found before the year 2000 and little availability of published literature on the case of developing countries. Additional field research could reveal scientific work that has not been reported in the literature.

Conclusions
This study analyzed the wider contexts of urban logistics measures based on economic development classes and geographical regions, including their significance in the transfer of urban freight policy. Policy transfer is a process involving multiple steps and many variables, among which the characteristics of the source and recipient city and the context within which the measure (package of measures) is applied in the source city are the cornerstones. Understanding the context in the source city can thus ease the appraisal and selection of measures for the recipient city.
The SLR method was used here for retrieval of a relevant and representative sample of published articles describing urban logistics measures, to ensure transparency in drawing objective conclusions. The analytical approach was based on a broad categorization of topics/measures into different economic development levels and geographical regions, to obtain a comprehensive global perspective with an adequate level of detail. Generalization of topics to cities located within the same economic development class and region (continent) can facilitate the transfer of best policy on urban logistics.
The results revealed that the economic development class and regional location of urban logistics topics assessed in the selected dataset were highly diverse. The economic development level and geographical location of a city were both found to have a strong association with the context in which measures were implemented. The contexts for the topics also reflected the nature of existing problems, which permitted appraisal of the suitability of measures for transfer to cities with similar challenges. Finding favorable contexts for measures in the recipient city is central to the success of the whole process.
Barriers and drivers to the transfer of measures were identified based on the contexts when evaluating the potential for transfer between two different exposure levels (developed and developing countries). In particular, institutional and physical barriers were found to be relevant to the transfer of urban logistics measures. Weak institutional set-up to formulate and regulate policy was identified as an institutional barrier and lack of adequate physical infrastructure provision as a physical barrier. Cities in developing economies should thus aim to reduce or counteract the influence of those barriers. Specifically, the measures (or package of measures) chosen should not require strong institutional control and follow-ups, to reduce the impact of institutional barriers, and should not require additional physical infrastructures, to reduce the impact of physical barriers.
The drivers depended on the specific contexts in the recipient city. In some cases, barriers to implementing one measure were a driver for implementing another measure. Identifying the specific enabler of measures in the recipient city can thus be of prime importance in the overall transfer process.
There has been little research about the transfer of urban logistics policies and the overall success in meeting the intended purpose in the recipient city, so this study makes an important novel contribution by providing a global picture of contexts in urban logistics measures. The results can also help to identify a broad set of barriers and drivers for cities in developing countries on measures originating from developed countries.

Supplementary Materials:
The following are available online at http://doi.org/10.5281/zenodo.4004037, Table S1: The complete list of retrieved and selected journal articles; Table S2: Full list of urban logistics measures contexts over economic classes.
Author Contributions: A.K.R. retrieved and selected the review articles, and conduct the analysis, and also wrote the article; G.G., L.T., and D.L. reviewed the article; All authors have read and agreed to the published version of the manuscript.
Funding: This work is supported by CBP-Ethiolog project (project no. NICHE-ETH-285) funded by Netherlands Initiative for Capacity Development in Higher Education (NICHE).

Conflicts of Interest:
The authors declare no conflict of interest. Table A1. Sustainable urban logistics (UL) topics identified in the selected dataset, with corresponding descriptions and the total number of articles in which they are mentioned.

Topic Description Total Number
Traffic congestion management related to UL operations The effect of traffic congestion on urban logistics operations and vice versa 6 Crowd-shipping Local people as carriers to pick up/drop off goods from automated parcel lockers or end-consumer locations, in dedicated or non-dedicated trips, to reduce environmental impact and possibly the cost of delivery   All measures involving payment of a toll for use of particular infrastructures, such as city centers, LTZ, bridges, and bypasses 2 Delivery robots/drones The use of drones or autonomous robots for urban deliveries, especially the last-mile segment 1 Table A2. General problems related to urban logistics in developing countries.

Summaries of Main Problems
Urban development and land use High population density, low level of infrastructure development, fragmented industry structure, and extensive informal sector [11,22]; mixed land use [43]; urban sprawl [44] Institutional set-up and role Poorly developed municipal and regulatory institutions [45]; limited institutional capacity [46]; institutional and legal framework ignored in the application [47]; lack of freight and related data [11] Policy and planning for urban logistics Lack of initiatives for urban freight [48]; poorly defined strategic policies and limited implementation [46] City traffic conditions Lower availability and quality of transport system [44]; weak traffic management [45]; extreme variety in transport modes and vehicle sizes; frequent breakdowns and accidents [11]; walking and cycling often unpleasant or unsafe [49] Freight infrastructures Infrastructure for walking and cycling often lacking [49]; lack of logistics infrastructure like loading/unloading areas (with the few existing being used for other purposes) [43,46,48]; parking and road design problems, and insufficient road network [43]; a narrow and often unpaved road, congested intersection, obstructed sidewalks and roadways, and uncontrolled access of heavy delivery trucks [50] Urban freight operations The existence of many small traditional retail stores (nano-stores) complicates the distribution of goods [48,51,52]; increase in truck movement, lack of coordination, and overloading of trucks [11]; higher empty returns increase the cost of trucking even for short distances [53]

Environmental impacts
Emissions from old vehicles [43]; high pollution from trucks using leaded gasoline [45]