https://orcid.org/0009-0006-6210-2602
Figures
Abstract
This study investigates the critical challenges associated with location selection for large-scale Agricultural Product Wholesale Markets (APWMs) under the traditional circulation model. It identifies and elaborates on the evolving characteristics of circulation stakeholders, supply chains, distribution channels, organizational structures, and external environments during the transition from traditional to modern circulation systems. In response to the demands of modern circulation, a comprehensive location selection evaluation framework is proposed, integrating five key criteria: location, planning, transportation, land use, and urban compatibility. Elastic and rigid evaluation standards are established according to the nature of each criterion. The framework innovatively integrates national territorial and spatial planning, road traffic planning, industrial development planning, and urban big data resources through Geographic Information System (GIS) technology, consolidating these into a unified database. To determine comprehensive weights for different functional types of APWMs, the normalized linear aggregation method is applied to combine weights derived from the Analytic Hierarchy Process (AHP) and the entropy weight method, enabling an analysis of correlation and contribution levels. Furthermore, this study introduces an innovative application of the Genetic Algorithm (GA), implemented in Python, to re-optimize the integration of subjective and objective weights through iterative computation until convergence, thereby enhancing the accuracy of comprehensive weight estimation and validating the location selection outcomes. A case study demonstrates the successful development of a five-phase location selection methodology—region screening, scope delineation, data analysis, weight optimization, and comprehensive evaluation—enabling both quantitative ranking and recommendation of candidate location and the optimal solution was ultimately selected from seven candidate schemes. This research provides practical guidance for location selection of large-scale APWMs within modern circulation contexts and offers methodological insights applicable to urban logistics planning and the siting of other large-scale infrastructure facilities.
Citation: Xin G, Tang Y (2026) A study of location selection for large agricultural wholesale markets under the perspective of modern circulation. PLoS One 21(4): e0345727. https://doi.org/10.1371/journal.pone.0345727
Editor: Tong Zhang, Southeast University, CHINA
Received: July 21, 2025; Accepted: March 9, 2026; Published: April 7, 2026
Copyright: © 2026 Xin, Tang. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Data Availability: In accordance with the Law of the People’s Republic of China on Guarding State Secrets, certain data related to national economic and social development are classified and therefore cannot be fully disclosed. To ensure transparency and reproducibility, the non-sensitive portion of the data has been made available in the Supporting information. For restricted data, a detailed description of the application procedure for access is provided. The territorial and spatial planning data, location data of large-scale infrastructure facilities such as airports, railway stations, and ports, dual evaluation data, urban development boundary data, road traffic planning data, location data of key municipal infrastructure, and satellite imagery used in this study are classified as sensitive data resources. These datasets are managed by third-party institutions. Access permissions may be requested from the Qingdao Natural Resources and Urban-Rural Planning Bureau when necessary. Contact information is available at http://zrzygh.qingdao.gov.cn/hdjl/202204/t20220402_5187517.shtml. Additionally, metro line data constitutes another sensitive data resource, for which access permissions can be obtained from the Qingdao Metro Group. Relevant contact details are accessible at http://www.qd-metro.com/. For your convenience, we have uploaded the Third Phase Construction Plan of Qingdao Urban Rail Transit (2021-2026) (Draft for Public Display) as supporting documents to the submission system. For the Qingdao benchmark land price inquiry system, please visit the website: http://218.58.72.174:8082/diit-web-landPriceMonitorIntnet/app/index.html.
Funding: The author(s) received no specific funding for this work.
Competing interests: The authors have declared that no competing interests exist.
Introduction
In China, the APWM is the primary channel for the circulation of agricultural products and occupies a central position in the overall system [1]. Statistical evidence indicates that over 80% of the city’s fresh products are sourced from this wholesale market [2]. The large-scale APWM functions as a critical element in the urban food supply chain, playing a pivotal role in various aspects such as price formation [3] circulation efficiency, food safety, agricultural industrialization, emergency protection [4], and more. The large-scale APWM generally refers to the first-level wholesale markets with an annual operation volume of 1 million tons or more up to over 3 million tons [5]. The 2023 No. 1 Central Document of China proposed the construction of a diversified food supply system and the establishment of a broad food perspective. Enhance the backbone network for agricultural product circulation by transforming and upgrading wholesale markets in production regions, distribution regions, and consumption regions, including: PWM, DWM and CWM. In 2024, the Ministry of Commerce, in conjunction with nine other departments of China, released the “Action Plan for Enhancing the Modern Commercial and Trade Circulation System and Promoting High-Quality Development of the Wholesale and Retail Sector.” The plan outlines the objective of establishing, by 2027, a modern commercial and trade circulation system that integrates domestic and international markets, connects urban and rural areas, links production with sales, and ensures efficient and smooth operations. Furthermore, the wholesale and retail sector is expected to accelerate its transformation, cultivate 100 key APWM, and continuously enhance its circulation organizational capabilities. The 2025 No. 1 Central Document of China proposes to expedite the development of a modern distribution network for agricultural products and agricultural supplies, while encouraging diverse entities to collaboratively construct supply chains. Establishing large-scale APWM is both urgent and highly significant for standardizing the agricultural product circulation system.
However, under the traditional circulation model, China’s agricultural products face several critical challenges: an excessive number of distribution stages, high logistics expenses, insufficient rates of cold chain storage and transportation, low levels of informatization and standardization, as well as imbalanced development attributed to a lack of modern trading platforms. The central issue lies in the location selection for APWM, which is associated with the following problems:
Inadequate location planning and insufficient comprehensive coverage: high-quality production regions, specialized cooperatives, and key consumption areas fail to establish full connectivity based on market attributes such as origin, sales, and distribution centers. This results in an increase in circulation costs.
Functional isolation and insufficient coordination: There is a lack of integration with relevant plans, such as territorial and spatial planning and industrial planning, to achieve mutual coordination with surrounding land functions. Additionally, the absence of corresponding specialized planning has resulted in disconnected links with the layout of urban warehousing, logistics, and distribution facilities, thereby trapping the extension of the industrial chain.
Inconvenient access and inefficient production-sales linkage: Many markets are situated far from main roads and deeply embedded within urban areas, leading to weak traffic diversion capabilities. Urban freight passage disrupts peace and tranquility. Cold chain storage and transportation connections are inadequate, hindering timely delivery. In China, post-harvest losses of agricultural products, particularly fresh vegetables and fruits, reach 20% to 30%, whereas in developed countries, these losses range from only 1% to 5% [6].
Dispersed layout and insufficient supporting infrastructure: The market is characterized by its fragmentation and limited size, with inadequate space for expansion. There is a paucity of facilities dedicated to trading, processing, inspection and quarantine, cold chain storage, logistics and distribution, research and development, and exhibition. This deficiency hinders the attainment of economies of scale and the facilitation of a shared economy.
Market-urban environment conflicts and mutual interference: Markets are increasingly surrounded by urban areas, exacerbating issues of a dirty, disorderly, and suboptimal trading environment [7]. These problems hinder the development and transformation of both the “market” and the “city,” necessitating urgent improvement and modernization.
Contemporary economic systems are characterised by the increasing volumes and flow of goods. The importance of geographical location is more prominent in the logistics process [8]. In location theory, location characteristics and efficiency are common focuses in planning and urban studies [9]. In location theory, under the influence of modern circulation, the connection between location and efficiency has become even closer. In addition to relatively efficient accessibility, the issue of fairness to service recipients in terms of location selection is also worthy of attention [10]. Just as the correct selection of the location for logistics centers is crucial for efficient trade and the economy [11]. The locational attributes of large-scale agricultural product wholesale markets are critical to the efficient functioning of the entire agricultural product distribution system.
Based on the aforementioned analysis, it is evident that whether constructing and nurturing new markets or relocating old ones, the scientific nature of location selection has emerged as the primary issue in resolving the contradiction between “market” and “city” and establishing a modern urban-rural agricultural product circulation system.
In 2023, the Chinese government released the “Notice on the Layout and Construction of Modern Circulation Strategic Pivot Cities,” designating 102 national modern circulation strategic pivot cities. The national “14th Five-Year Plan for the Construction of a Modern Circulation System” emphasizes accelerating the development of a modern circulation system, fostering innovative circulation organizations and business models, and promoting seamless integration across upstream and downstream sectors, production, supply, sales, and domestic and international trade. Modern circulation connects production with consumption through an extensive supply chain network and links supply and demand via more precise and efficient logistics and distribution systems. The circulation of agricultural products has transitioned from traditional models to modern ones, resulting in significant transformations in transactional relationships and logistics organization models [12–14]. This shift from traditional to modern circulation introduces numerous new characteristics and provides a novel framework for addressing locaion selection challenges under the traditional circulation paradigm.The research aims to develop a comprehensive locaion selection evaluation system in response to the development trends and demands of modern circulation. It focuses on selecting appropriate and operable evaluation methods, integrating them with territorial and spatial planning and industrial planning, to achieve quantitative assessments for the location selection of large-scale APWM including PWM, DWM and CWM. This study will support regions, such as China, which have an urgent need to establish APWM system, in designing effective location selection evaluation approaches and improving the efficiency of modern circulation systems.
Literature review
Modern circulation
In a previous study, Cadilhon et al. [15]. examined the efficacy of contemporary marketing channels in developing countries and their effect on traditional methods. Oparebea Boateng et al. [16] pointing out the tendency of supermarket purchasing to distribution centers, specialised/dedicated wholesalers, preferred supplier programmes in the context of the shift from traditional to modern distribution models. Xu [17] proposes a contemporary circulation model as a consequence of the modern logistics industry’s enhanced productivity. This model integrates “new logistics” and “new retail” in a symbiotic relationship, giving rise to innovative forms of logistics such as instant and agile, intelligent and autonomous logistics. Logistics is a manifestation of circulation, which has also undergone a major transformation under the influence of modern distribution trends. Danyluk [18] proposed logistics revolution has altered sociospatial processes at multiple sites along the supply chain. Characterized by a high degree flexibility and the dense interdigitation of the spaces of production and circulation. Cichosz [19] proposed determinants of innovative business models in the digital era Including Connectivity, Cooperation, Integration, Adaptiveness. In addition, Cichosz et al. [20] listed building information platforms, employee and partner engagement, and enhancing the customer experience for logistics companies under Digital transformation as the success factors for the transformation. The impact of modern circulation on the industry chain is also worthy of attention. Hu [21] proposed that the modern circulation industry should be integrated with related industries and innovation, vigorously developing the integration of urban and rural production, supply, and marketing of big data platforms, as well as the comprehensive integration of urban and rural production, supply, and marketing of various enterprises and farmers in the chain. Chen [22] proposed modern circulation is defined as a comprehensive, open, and dynamic process that encompasses the entire cycle of production, distribution, circulation, and consumption.
Modern circulation is embedded in urban mobility and is closely related to traffic dynamics, governance and policy integration mechanisms, and community equity [23]. When selecting locations for large-scale agricultural product wholesale markets, it is essential to comprehensively consider their roles within the circulation network and the various stakeholders associated with them. Additionally, leveraging information technology and efficient logistics infrastructure to establish interconnected systems can effectively enhance circulation efficiency.
Location selection of wholesale market and logistics center
Large-scale APWM are important logistics, warehousing and distribution infrastructure. There have been numerous studies on the location selection of similar facilities such as logistics centers and wholesale markets. Aiming at the five aspects of the problems faced by traditional distribution in the previous study, the criteria influencing the location of wholesale markets or logistics centers for agricultural products listed in the literature were categorized and summarized in Table 1.
For the method of selecting the location of agricultural markets or logistics centers, from the point of view of optimal logistical efficiency Özmen et al. [34], Xiao [24] selected key criteria were chosen for the locaion selection study；Raimbekov et al. [40] used AHP and centroid method, Muerza et al. [32] compared ANP algorithms, Kieu et al. [39] used hierarchical analysis (SF-AHP) and combined compromise solution (CoCoSo) algorithms, Pham et al. [35] used Delphi TOPSIS method, and the above studies were conducted at national and provincial regional level for locaion selection. Zhang [30] used entropy weight TOPSIS combined with GIS technology, Giuffrida et al. [41] used GIS combined with the Delphi method to conduct a study on the location of agricultural markets or logistics centers at the urban level. While for subjective or objective quantitative analysis has its own focus, Uyanık et al. [29] conducted a literature review of relevant locaion selection methods, and in China relevant normative standards, policy documents also do not have a detailed agreement or guidance on location selection evaluation methods.
Based on the aforementioned analysis, the location selection scope primarily focuses on urban areas, with an emphasis on municipal regions. The locaion selection methodology should not only encompass qualitative approaches but also integrate quantitative techniques by leveraging GIS and big data resources to analyze the spatial attributes of the city. The foundation for location selection must consolidate the outcomes of territorial and spatial planning and industrial planning, ensuring a forward-looking and operational framework. More importantly, it is essential to integrate the features of modern circulation by forming a comprehensive distribution chain. Through a combined subjective and objective analytical approach, full-criteria quantification should be conducted to enhance the scientific rigor of locaion selection research for markets with varying functional attributes.
Urban planning and accessibility impacts.
Large-scale APWMs play a significant role in urban planning and, in turn, exert a substantial influence on urban development strategies. Charters-Gabanek et al. [42] review presents a broad discussion of the links between the logistics industry and a series of urban planning concerns, develop a framework around the forces shaping the supply of, and demand for, land for logistics to engage with current conversations in urban studies on the production of urban spaces. Kin et al. [43] put forward cities can support the (re)integration of logistics facilities in urban areas to facilitate and enable the shift to an efficient urban logistics system. Develop best practices on how to address the integration of urban logistics facilities for cities. Wang et al. [44] use the real data of Beijing analysis of the accessibility impacts of logistics suburbanization, highlights the potential for freight accessibility in urban freight planning. Chen et al. [45] By leveraging existing urban infrastructure, a multi-criteria evaluation framework integrating the AHP and the TOPSIS was developed to support the location selection of urban logistics centers. Furthermore, in the research on the impact of transportation accessibility on vulnerable groups, Abdulrazaq et al. [46] established a priority evaluation framework for vulnerable road users and developed a spatio-temporal cube analysis approach by integrating XGBoost to assess the impact of emerging factors on traffic accessibility.
Methodology
Location selection strategy for modern circulation
Through a comprehensive review of modern circulation-related literature and an analysis of relevant policies, it is evident that the development of agricultural product wholesale markets must align with the requirements of modern circulation systems. This involves integrating multi-level resources to create favorable conditions for establishing a modern circulation platform that encompasses production, supply, sales, processing, cold chains, warehousing, distribution, and information interconnectivity. Based on the five key issues inherent in traditional circulation models, this study specifically examines the new characteristics exhibited by circulation entities, circulation chains, circulation channels, circulation organizations, and circulation environments during the transition from traditional to modern circulation. Additionally, problem-oriented considerations for the location selection of large-scale agricultural product wholesale markets are systematically outlined in Table 2.
The establishment of the location selection evaluation system for new circulation
In light of the aforementioned five major issues in traditional circulation and the transition from the five characteristics of traditional circulation to modern circulation, the location orientation is transformed into an evaluable influencing criterion. By integrating existing research, location criteria are consolidated, supplemented, refined, or quantified. Consequently, a location evaluation system comprising five primary influencing criteria is constructed to fulfill the requirements of modern circulation.
Location synergy (L) -- integration of multiple market systems.
In light of the location coordination issue, existing studies have investigated the correlation between production and consumption locations [25,26,34] as well as the allocation of resources in central urban areas [29,30,34]. Within the framework of modern circulation systems, market selection should emphasize hierarchical refinement and the linkage relationships among markets at various levels. This involves integrating the “three vertical and three horizontal” backbone network for national agricultural product circulation [51] with a three-tiered agricultural product origin market system [7], which is structured with the national level as the leader, regional level as the node, and field markets as the foundation. Consequently, this forms a circulation pattern characterized by orderly linkage, with national or backbone markets and regional markets serving as primary nodes as show in (Fig 1).
Caption credit: This figure was created by the authors using Adobe Illustrator.
At the micro level, modern circulation relies on multi-channel interactions from the production end to the consumption end, guided by an overarching smart platform. This has given rise to numerous modern circulation pathways as show in (Fig 2), such as “agricultural wholesale and retail connection,” “agricultural supermarket connection,” “agricultural community connection,” and “order-based agriculture.”
Collection of multiple business modules include: E-commerce, cold chain, Storage, processing, distribution module. Caption credit: This figure was created by the authors using Adobe Illustrator.
On the one hand, the regional layout is well-coordinated (L1). By considering the distinct functional attributes of production, consumption, and distribution markets, the market entity is further categorized into: external city markets (L1-1), advantageous production areas within the city (L1-2), lower-level origin market (L1-3), lower-level consumption markets (L1-4), farmers’ markets (L1-5), and major supermarkets (L1-6). This classification ensures comprehensive coverage of connections with various market entities and facilitates a comparative analysis of circulation costs. On the other hand, in light of the integration of modern circulation with human resources, market clusters, and public service facilities within the central urban area, and adhering to compliance requirements such as “Planning for the One-Hour Fresh Agricultural Products Circulation Circle around the Capital” and “Management Specifications for Instant Retail Operations (DB42/T 2325-2024),” attention is given to the one-hour delivery radius in relation to the central urban area (L2), which serves as a rigid criterion for locaion selection.
Planning coordination (P) – Interconnecting multi-functional spaces.
For the functional planning problem, studies have primarily focused on “proximity to organized industrial zones,” the scope of urban development in terms of “urban development planning” [34], and industrial policy [7]. Modern circulation emphasizes the establishment and expansion of a complete industrial chain. Locaion selection should integrate the requirements of various levels of territorial and spatial planning and industrial planning regarding the layout of industrial functional spaces, thereby enabling comprehensive overall planning and constructing an integrated network.
Link up the territorial and spatial planning system and coordinate the functional layout of land use (P1): To align with the content and requirements of the comprehensive territorial and spatial planning, location selection must consider its dual role as both a transit station for agricultural products outside the city area and a supply and security station for urban agricultural products. Based on spatial-temporal evolution research, such locations are predominantly distributed in inner or outer urban areas [52]. This guides location selection to adhere to the suburbanization principle of “close to the city but not within it,” as exemplified by cities listed in Table 3. A rigid principle for location selection is that the location must be situated near or within the urban development boundary. Regarding land use, wholesale market land falls under commercial land according to land use standard (Guide to classification of land and sea use for territorial space survey, planning and use control) in China. In conjunction with existing land function plans, it is imperative not to occupy residential, public service facilities, special-purpose land, scenic spots, or other designated areas as a fundamental planning principle. Looking ahead, considering the integration of transaction, incubation, processing, and logistics industry chains for large-scale agricultural product wholesale markets under modern circulation systems, priority should be given to locating industrial land such as commercial service, storage, and industrial land. Additionally, by combining industrial control line demarcations, extract big data resources such as POI related to agricultural products or food processing, logistics warehousing, and cold chain enterprises. Select relatively clustered industrial development zones, compare the correlation degree of candidate sites, and incorporate elastic evaluation indicators. In the later stages, locaion selection results will be validated and adjusted through detailed or specialized planning as show in (Fig 3).
LSES, Location selection evaluation system. Caption credit: This figure was created by the authors using Adobe Illustrator 2023.
Integrate the industrial spatial layout (P2): Link existing plans, and consolidate industrial development strategies with specialized plans such as agricultural modernization, food industrialization, cold chain storage and distribution, and logistics development. Evaluate the correlation of relevant industrial parks, processing parks, logistics parks, and other key industrial development zones as an elastic evaluation criterion for locaion selection. Simultaneously, transmit the outcomes of the location selection process to industrial special planning and project planning as show in (Fig 3).
Transportation accessibility (T) -- Expanding multi-dimensional circulation channels.
In addressing the issue of transportation connectivity, modern agricultural product circulation relies on efficient information interconnection to rapidly establish logistics connections among production, supply, and sales. Concurrently, various e-commerce models have emerged, such as “fresh food e-commerce + cold chain home delivery,” “central kitchen + cold chain food delivery,” O2O (online-to-offline), and C2P (consumer-to-producer) [14]. Modern circulation tends to customize circulation plans for different channels, select optimal circulation paths as shown in (Fig 2), and facilitate access to multi-level transportation networks both domestically and internationally. In conjunction with existing research on traffic impact [29,30] five key traffic impact factors are established: intercity expressways (T1), urban transportation (T2), freight railway facilities (T3), port and wharf facilities (T4), and airport facilities (T5). Emphasis is placed on the seamless integration of intermodal transport and personalized transportation demands within modern circulation systems. A comprehensive quantitative analysis is conducted to balance the rigidity and elasticity of all criteria involved.
Given that over 75% of agricultural products are transported via road networks [40], where expressway accessibility is a critical factor influencing the efficiency of commodity logistics [53]. Efforts should be made to minimize the distance between markets and expressways, enhance operational efficiency, and reduce urban impacts. Drawing from investigations of major agricultural markets, such as Rungis in France, Toyosu market in Japan, and Xinfadi market in Beijing, a maximum traffic connection time of 5 minutes to expressway entrances and exits is recommended.
Considering the increasing proportion of air transport for agricultural products, the low-altitude economy has established a modern circulation pathway for agricultural product transportation. Air cargo not only enhances the value of transported goods but also broadens the range of circulation participants [54]. This highlights the importance of connecting locaion selection with various levels of airports. According to research findings, logistics facilities strongly associated with airports are typically located within 20 kilometers of an airport [55,56]. A driving limit of 30 minutes from the airport is defined based on average freight speed of 40 km/h [57]. and the control time for minimizing conventional transportation losses of fresh goods entering and leaving the port. The “urban transport (T2)” aspect considers its significant circulation and traffic demands, emphasizing direct connections with urban trunk roads or urban expressway (T2-1). These three points serve as rigid principles. Additionally, in response to the accessibility needs of diverse groups in the agricultural batch market under modern circulation, public transport accessibility (T2-2) evaluation indicators are introduced to comprehensively assess the overall transportation conditions of the selected location.
Land suitability (S) – Supporting the whole chain circulation service.
In response to the layout of supporting issues, the agricultural product wholesale market within the modern circulation system tends to establish a large-scale intelligent circulation platform that integrates the entire industry chain, including live e-commerce, processing and warehousing, cold chain logistics, centralized quarantine, expos, research and development incubation, recycling and reuse. This platform aims to provide comprehensive, one-stop, and highly efficient services. To ensure the successful implementation of such a platform, several critical criteria must be considered: sufficient scale and expandability (S1), strong industrial foundation (S2), reasonable land costs (S3), favorable land consolidation conditions (S4), and an appropriate environment (S5), including suitable construction site conditions (S5-1) and landscape features that facilitate market integration into the city (F5-2). Among these criteria, “Land use scale and expandability (S1)” is derived from domestic large-scale agricultural market research, referencing the size of urban-level wholesale markets exceeding 66.67 hectares [58], as well as the construction area and floor area ratio standards for large-scale APWM with an annual transaction volume of 300 tons or more [5]. Based on this, a recommended rigid locaion selection value of at least 65 hectares is established. Integrate the planning of suitable construction areas, as determined by the “two evaluation” of territorial and spatial planning, into the section “Site Conditions (S5-1).” In compliance with normative requirements (Vertical Planning Code for Urban and Rural Construction Land (CJJ83−2016) published by China), strive to ensure that the natural slope of the land does not exceed 15% wherever possible, thereby minimizing construction impacts resulting from adverse conditions such as natural disasters. Considering the modernization of the business format in APWM and the need to enhance the quality of the construction environment, integrate with the territorial and spatial planning system. The planned park green space within an 800-meter buffer zone that can be reached by walking at a comfortable adult walking speed of approximately 0.9 m/s [59] is correlated with the “landscape conditions (S5-2)” and integrated into the locaion selection evaluation framework. This promotes the transformation of the market into a new driver for characteristic cultural tourism and urban vitality enhancement.
City friendly (F) – Standardize the operational sequence of multiple types.
In response to the issue of urban interference, modern circulation methods have become increasingly diverse and adaptable. These methods not only ensure timely delivery to downstream markets or direct consumers but also contribute to maintaining a peaceful and orderly urban environment. To achieve modern circulation objectives, it is necessary to introduce concepts and standardized guidelines that facilitate quantifiable and expanded evaluation indicators. Drawing on the principles outlined in the “Technical Guide for Community Living Circle Planning” and the “Guide for the Construction of a Quarter-Hour Convenient Living Circle in Cities,” part of the APWM could be transformed into a 15-minute convenient distribution system. This system would enable prompt distribution while avoiding homogeneous competition. Additionally, a rigid principle should be established, ensuring a minimum distance of more than 15 minutes between existing markets (F1-1). Such measures aim to establish a regionally responsive, flexible, and convenient supply system. Under modern circulation, it is essential to emphasize the integration of epidemic prevention with peacetime and emergency preparedness, ensuring the smooth operation of the emergency support system. Additionally, the evaluation of the efficiency of urban logistics and emergency support facilities (F1-2) should be enhanced, and the agricultural product circulation system should be incorporated into the framework of urban safety and resilience guarantees. To foster a peaceful and orderly urban living environment while preventing pollution, it is recommended that the distance between residential areas and large public service facilities such as education and healthcare (F2-1), as well as toxic and hazardous pollution sources or dangerous sites (F2-2), be maintained at more than 1 km [60,61] as a rigid condition. This ensures minimal interference and impact on residential areas and large public service facilities, while simultaneously safeguarding the operational environment of agricultural markets. These measures collectively reflect the goal of fostering city-city amity under the context of modern circulation.
To ensure alignment between policies and established standards, Table 4 outlines the correspondence among policies and related research findings across multiple regions, thereby facilitating a clearer understanding of the relationship between the evaluation criteria adopted in this study and existing scholarly work.
To summarize, a location selection evaluation system has been established, comprising 5 primary location selection criteria, 16 secondary location selection criteria, and 25 sub-classification criteria. The overall criteria classification, along with other elastic evaluation principles, is elaborated in Table 5. Transportation convenience and cost are important criteria to be considered in location selection [29], while time cost often has a greater impact on logistics efficiency [62]. Therefore, in order to ensure fairness, the ADT to one or more analytical destinations is adopted as a key metric in specific elements for comparing circulation benefits.
Evaluation methodology
Based on the integration of rigid and elastic approaches within the hierarchical framework of the location selection evaluation system in Table 5, various location selection methodologies are systematically categorized. Among these, the AHP has emerged as one of the most widely adopted techniques for logistics location selection [29]. AHP facilitates the organization of expert input from diverse fields, enabling the establishment of a layered evaluation matrix that aligns with the multi-level assessment system for location selection criteria.
Additionally, the entropy weight method provides an objective approach to weight calculation by leveraging data sourced from GIS, big data platforms, and other relevant sources. Consequently, the combination of AHP and the entropy weight method is employed for location selection evaluation. This integrated approach not only ensures the accurate computation of weights based on objective data but also accommodates expert adjustments to criterion weights in accordance with the specific requirements of different cities and markets.
According to the classification and evaluation criteria presented in Table 5, the location selection evaluation system is implemented through a structured process that includes region selection, scope definition, weight determination, data analysis, and comprehensive evaluation. This process facilitates the location selection assessment for different types of large-scale APWM, including PWM, DWM, and CWM, as illustrated in (Fig 4).
Caption credit: This figure was created by the authors using Adobe Illustrator 2023.
Data and results – the case of Qingdao city
Establishment of location selection data base
Qingdao is situated in the Huang-Huai-Hai Plain National Quality Agricultural Production Area and serves as an international comprehensive transportation hub city, strategically positioned within the “three vertical and three horizontal” agricultural products circulation backbone network, which is also a modern circulation pivot city determined by the state. Based on the location selection criteria and principles outlined in Table 5, take the municipal boundary of Qingdao as the research scope, this study integrates the Territorial Spatial Planning of Qingdao (2021–2035), the Comprehensive Transportation System Plan of Qingdao (2021–2035), the territorial spatial planning of other districts in Qingdao City, relevant 14th Five-Year Plans, and other specialized plans of Qingdao City. Open-source data such as Open Street Map (OSM), DEM data, and big data POI are extracted to establish the necessary data resources for evaluation. The speed limit requirement of 100 km/h for freight vehicles on intercity expressways in China, combined with the statistics provided by the Bureau of Transportation Statistics of the U.S Department of Transportation, the average speed of trucks on interstate highways in 2019 ranged approximately from 48 to 62 miles per hour [53,63] (77–100 kilometers per hour), take the average value 90 km/h as the simulated average driving speed of the intercity expressway. Urban expressway, trunk roads, and secondary trunk roads, given the traffic congestion typically experienced in urban areas, reference should be made to the “Code for Design of Urban Road Engineering (CJJ37-2012),” adopting the minimum design speed as specified, set the average driving speeds of 60 km/h, 40 km/h, and 30 km/h.
As for the location data of the existing market, it mainly comes from the Qingdao Municipal Bureau of Commerce, the Qingdao Municipal Market Supervision Administration commissioned the Qingdao Municipal Urban Planning and Design Institute to compile the Qingdao Municipal Commercial Network Special Plan (2020–2035) and the Qingdao Agricultural Market Special Plan (2021–2035) respectively. According to the official website of the market and the above plan, the three types of large-scale APWM including PWM, CWM, and DWM within the city are identified as base markets and incorporated into the established database as show in (Fig 5).
Caption credit: This figure was created by the authors using ArcGIS 10.8.1, Adobe Illustrator 2023.
Region selection analysis
The stringent conditions associated with L2, P1, T1, T5, S5, and F1-1 were computed using GIS. Specifically, vehicle travel isochrone service area calculations were performed for L2, T1, T5, and F1-1. The results of L2, P1, T1, T5, and S5 were aggregated, while the analysis range of F1-1 was excluded. This process identified a total of seven potential location selection areas: A, B, C, D, E, F, and G as shown in (Fig 6).
(a) L2: One-hour isochrone distance from the central urban area; (b) P1: Analysis of Urban Development Boundaries; (c) T1: 5-minute isochrone distance from entrance and exit of expressways; (d) T5: 30-minute isochrone distance from airport; (e) S5-1: Analysis of Suitable construction region; (f) F1-1: 15-minute isochrone distance from current different types of wholesale markets; (g) Analysis results of superimposed exclusion under the rigid principle of six criteria. Caption credit: This figure was created by the authors using ArcGIS 10.8.1, Adobe Illustrator 2023.
Subsequently, adjacent areas within the 15-minute isochrone of the existing market (F1-1) were disregarded, areas B and F are excluded, resulting in the final screening of five areas: A, C, D, E, and G as shown in (Fig 7).
Caption credit: This figure was created by the authors using ArcGIS 10.8.1, Adobe Illustrator 2023.
Scope definition analysis
For the areas encompassed within or adjacent to the proposed location selection scope, the stringent criteria associated with operational criteria T2, S1, and F2 were applied to screen the proposed location selection scope. Following this process, areas D, E, and F were excluded from the location selection due to NIMBY concerns related to residential and public service facilities in (Fig 8). Since area F has already been excluded in the “Region selection analysis”, therefore areas A, C, and G each identified one suitable location selection scope, as outlined in (Fig 9).
Caption credit: This figure was created by the authors using ArcGIS 10.8.1, Adobe Illustrator 2023.
(a) Analysis of the surrounding region of location A1; (b) Analysis of the surrounding region of location C1; (c) Analysis of the surrounding region of location G1. The coordinate system uses the CGCS2000 (Gauss-Kruger 3° Zone 40) coordinates, where X represents northward coordinate and Y represents eastward coordinate, including zone number. Caption credit: This figure was created by the authors using ArcGIS 10.8.1, Adobe Illustrator 2023.
The calculation of the elastic principle criteria
With reference to 13 relevant plans and big data resources from Qingdao City, data computation or statistical analysis was performed for the evaluation indicators of the three proposed location selection scope.
GIS technology was utilized to construct an OD cost matrix to calculate the average travel time for the elastic criteria associated with L1, P1, P2, T3, T4, T5, and F1-2. The requirements of Qingdao and cities along the Yellow River Basin agricultural products and agricultural materials circulation corridor to build agricultural products circulation platform [64], combined with Qingdao coastal location and road circulation path, the L1-1 criterion is derived by computing the average driving time from the location to the intersection points where the municipal boundary line meets each intercity expressway. In the calculation of the P1 criterion, given the average vehicle speed of 20.7 km/h [65] during peak hours on weekdays in Qingdao, it takes approximately 10 minutes to cover a radius of roughly 3 kilometers. Taking into account the principle of proximity-based collaboration, this estimate is adopted as the permissible coverage area for the relevant industrial cluster. For L2, T1, T5, and F1-1, the nearest facility point running time was determined. The commuting time for T2-2 is estimated by calculating the travel distances covered via subway and bus, combined with the average speeds associated with each mode of transportation. The subway network was constructed based on the government-approved report [66]，with an average operating speed of 44.83 km/h derived from the operator’s official statistical report [67]，Bus routes were verified using Baidu Maps Navigation, and the average bus speed was determined based on the peak-hour average speed of 16 km/h specified in the government-issued 2025 annual transportation plan [68]. The peak-hour speed is adopted to reflect the concentration of major trading activities in large wholesale markets during the morning rush hour, thereby fully accounting for the traffic congestion likely to occur during their peak operating periods. For T2-1, S1, S2, and F2, corresponding data were used for map-based measurements. The impedance values and their parameter sources used in the commuting evaluation criteria are summarized and explained in Table 6. All GIS-based parameter calculations were conducted within the CGCS2000 coordinate system Fig 10–12. The analysis of certain criteria is presented in Figs 10–12. Additional calculation principles are elaborated in Table 7.
(a) L1-1: ADT at the intersection where expressways meet the municipal boundary; (b) L1-2: ADT with the advantageous production areas; (c) L1-3: ADT with the secondary production market; (d) L1-4: ADT with the central farmers’ market; (e) L1-5: ADT with the community farmers’ market; (f) L1-6: ADT with the major supermarket. Caption credit: This figure was created by the authors using ArcGIS 10.8.1, Adobe Illustrator 2023.
(a) L2: The shortest driving time to the city center; (b) P1: ADT of the food processing and cold chain logistics industrial concentration region in the territorial and spatial planning; (c) P2: ADT with the agricultural product processing park planned for the 14th Five-Year Plan. Caption credit: This figure was created by the authors using ArcGIS 10.8.1, Adobe Illustrator 2023.
(a) T1: The driving time to the nearest expressway entrance and exit; (b) T3: ADT with railway freight stations; (c) T4: ADT with major port facilities; (d) T5: The driving time to the nearest airport; (e) F1-1: The nearest large-scale APWM driving time; (f) F1-2: ADT with the logistics support facilities. LSC, Logistics Support Center. Caption credit: This figure was created by the authors using ArcGIS 10.8.1, Adobe Illustrator 2023.
Determination of comprehensive weight
Weight analysis based on AHP method.
Based on the pairwise comparison matrices of the criteria in Table 8, a panel of experts with diverse professional backgrounds—spanning government management, operational investment, cost control, industrial planning, territorial and spatial planning, architectural design, and potential merchants—was convened to assign weights to the 16 types of secondary criteria listed in Table 5, and allocate values to the sub-classification criteria under some secondary criteria based on the specific requirements of different market types, including production areas, consumption areas, and distribution areas, to construct a comprehensive matrix, refer to S1 Table, All individuals participating in the survey were informed of the research objectives and provided oral consent prior to participation. To protect privacy, we have anonymized all the experts. Consistency tests were performed separately using SPSSAU, with the consistency ratio (CR < 0.1). The effective matrix was then summarized using the geometric mean method, and the AHP weight values for the secondary classification and its sub-classification judgment matrix were calculated by inputting the data into the SPSSAU system (resulting CR = 0.031 < 0.1). Finally, pairwise multiplication was conducted to calculate the AHP weight values for all criteria , as presented in Table 9.
Weight analysis based on entropy weight method.
A judgment matrix of m schemes and n evaluation indicators is constructed based on the data in Table 7, and positive and negative dimensional processing is carried out according to their relevant attributes, and the processed data is entered into the SPSSAU system for analysis. The steps are as follows:
First, the evaluation matrix X is constructed as below:
where I=1, 2, …, n； j=1, 2, …, m.
Then the data was standardized and the forward indicators were calculated, as shown in Equation 2. The negative indicator is calculated, as shown in Equation 3. In order to avoid the situation that some data after standardization has a low value, H = 0.01 is taken for translation processing as shown in Equation 4.
Where is the standardized processing of the positive and negative vectors to avoid the influence of different measurement units of the elements, and H is used as a translation vector to prevent the occurrence of zero values.
Finally, the SPSSAU system is used to calculate the entropy weight in Table 9.
Comprehensive weight calculation and analysis of calculation results.
The comprehensive weight of AHP+ entropy weight method in Table 9 is calculated using Equation 5. Meanwhile, the standardized values calculated by Equation 2,3 are used to obtain the comprehensive scores of different evaluation criteria by using linear weighting method calculated as Equation 6, which is obtained in Table 10.
Where denote the AHP weight for criterion j and
denote the entropy weight; We calculate the comprehensive weight
by summing the products of the two types of weights and normalizing them.
According to the calculation, A1 location selection scheme is obviously superior to other schemes, especially in the layout of the production market, followed by the distribution market and sales market. G1 was the second choice after A1, while C1 had the lowest score. To analyze the results more clearly, this time the partial comprehensive method was used to analyze the contribution rate of the scores under the market weights of each type, in order to determine the contribution value of each element to the final score. Equation 7 represents the contribution value of the i criterion, and Equation 8 is the contribution degree ratio of the i criterion.
Where, denotes the difference in standardized scores
under the i-th factor evaluation between two locations, and
represents the corresponding comprehensive weight.
Where, denotes the sum of the differences in standardized scores
across all factor evaluations between two locations. A positive
value indicates the degree to which the criterion contributes to the location’s score, whereas a negative value reflects the extent to which it negatively impacts the score.
Given that A1 and G1 represent the optimal and suboptimal location alternatives, respectively, a comparative analysis was conducted to examine the influencing factors between these two schemes. The results under the comprehensive weight of the origin market are presented in Table 11, while the corresponding analytical outcomes under all attribute weights are illustrated in Figs 13–15.
Caption credit: This figure was created by the authors using OriginPro 2024b.
Caption credit: This figure was created by the authors using OriginPro 2024b.
Based on the analysis results, under the three weighting scenarios corresponding to origin markets, distribution centers, and destination markets, the three criteria contributing most to A1’s higher score relative to G1 are T5, P1, and T1, while those exerting the greatest negative impact are L2, F1-1, and T2-1. This indicates that A1 holds significant advantages in proximity to airport facilities, planned concentrated industrial development zones, and expressway exits, whereas G1 demonstrates comparative strengths in closeness to the city center, maintain an appropriate distance from existing markets., and access to major urban roads.
Analysis of the relevance of criteria and concentration of weights.
As the location selection process has resulted in three candidate areas following a stepwise screening procedure, and considering the limitation in sample size (n = 3), the spatial correlation among location selection criteria based on commuting time and the OD cost matrix is assessed using the Spearman rank correlation coefficient, as defined in Equation 9. The potential risk is quantified by the degree of weight concentration. Based on the analysis of the original data in Table 7, the performance of each criterion is ranked according to relative superiority in Table 12, where higher values of positive-direction indicators indicate better performance, while lower values of negative-direction indicators are preferable. Rank 1 denotes the best performance, rank 2 the second best, and rank 3 the worst.
Where, the Spearman rank correlation coefficient primarily reflects the rank-based association within the sample, rather than the statistical significance typical of large samples; n = 3 denotes the sample size, and
represents the difference in ranks.
The feature matrix derived from the results was processed using the Spearman rank correlation to compute the correlation coefficients among the various evaluation criteria, and a corresponding heatmap is presented in Fig 16.
* indicates that the P value ≤ 0.05. Caption credit: This figure was created by the authors using OriginPro 2024b.
Through analysis, it is evident that L1 exhibits a significant positive correlation with T2-2, T3, and F1-2; L2 with P1, P2, T1, and T5; P1 with P2, T1, and T5; P2 with T1 and T5; T1 with T5; T2-2 with T3 and F1-2; and T3 with F1-2. Additionally, F1-1 shows a significant negative correlation with L2, P1, P2, T1, and T5. The average criteria exhibiting higher correlations demonstrate varying degrees of weight concentration, as detailed in Table 13. These patterns are closely associated with the traffic-related geographical characteristics considered during the location selection process. The corresponding heatmap is presented in (Fig 16).
Sensitivity analysis of partial elasticity criteria
During the process of determining objective weights, the average speed for each road network class used in the OD cost matrix is established based on multi-year statistical data and in conjunction with road design standards, ensuring relative stability. However, the metro speed, bus speed, and peak-hour motor vehicle operating speed employed in the T2-2 and P1 evaluation criteria are derived from specific-year statistics or planning projections, exhibiting a certain degree of variability. These factors directly influence the OD impedance associated with the T2-2 criterion and the OD path under the P1 criterion. Therefore, a sensitivity analysis is conducted on these three types of influencing factors. By assuming adjustments of ±10% for metro speed, ± 20% for bus speed, and an increase or decrease of 5 minutes in commuting time within the 10-minute industrial cluster radius, a perturbation analysis is performed on the relevant parameters.
During the perturbation analysis of the P1 value, with commuting radius durations set at 5, 10, and 15 minutes, the corresponding commuting radii were converted to 1.7 km, 3.0 km, and 5.0 km, respectively. The GIS point density analysis function was applied to identify functional aggregation areas within these radii where the number of POIs for food companies and cold storage facilities exceeded 10. These identified areas were subsequently overlaid with the industrial development zones defined by national land space planning. The centroid of each overlapping area was designated as the destination point of the respective industrial aggregation zone. Under the three commuting time thresholds, the numbers of destination points meeting the criteria were 46, 84, and 114, respectively. The average commuting distance between the proposed market location and these destination points was calculated using the OD cost matrix (Fig 17), from which the statistical values for the P1 location selection criterion were derived.
The calculation and statistical results of the relevant criteria after the disturbance of the above elements are shown in Table 14.
Following the perturbation, the calculated comprehensive weight values for P1 and T2-2 are presented in Table 15. It can be observed that under a ± 5-minute variation in the industrial cluster commuting radius, the comprehensive weight changes within the range of −40% to −43%. This change is primarily attributed to the altered average commuting time resulting from shifts in destination points within the industrial concentration area, which reduces the comprehensive weight of this criterion. In contrast, perturbations in bus and subway average speeds lead to proportional changes in commuting time for T2-2, without affecting the corresponding weight coefficient.
When the weight of P1 varies, the comprehensive scores of the three location selection alternatives exhibit only minor fluctuations, with no impact on the overall ranking. The comprehensive scores and corresponding change rates for each site under different P1 market weight scenarios are presented in Table 16.
Comprehensive weight optimization based on GA
After the conventional application of the aforementioned methods to calculate comprehensive weights, it is recognized that not all indicators have been subjected to perturbation, and the correlation and weight concentration among certain indicators are taken into account. Furthermore, in multi-criteria evaluation, subjective weighting methods (e.g., AHP) rely on expert judgment but are susceptible to subjective bias, whereas objective weighting methods (e.g., the entropy weight method) depend on data characteristics yet exhibit limited reliability in small-sample settings. GA has been employed in several studies for location selection and optimization of transportation and logistics facilities, yielding favorable outcomes [69–72].For the high-dimensional, small-sample scenario of “25 evaluation indicators and 3 samples” addressed in this study, a GA-optimized AHP–CV (coefficient of variation) integrated weighting approach is proposed to enhance the validation and optimization of comprehensive weights for the three market types. In this method, AHP-derived weights reflect subjective expertise, CV-derived weights capture the objective variability within limited samples, and the two are adaptively combined through GA global optimization, thereby balancing theoretical rationality with data-driven adaptability in the evaluation process.
Subjective weights and objective weights
The AHP weights were determined through the pairwise comparison matrix derived from expert assessments of indicator importance, as presented in Table 9. This study adopts the AHP weight vector that has passed the consistency test (CR < 0.1).
Satisfy and
Where represents the AHP weight of the i-th indicator.
For the small-sample scenario involving three samples, the CV is adopted as the dispersion measure to address potential objective weight distortion, replacing the entropy weight method. The calculation formula is presented in Equation 10.
Where, represents the sample mean of the i-th indicator, and
is the sample standard deviation (with a degree of freedom of 1, suitable for small samples).
The objective weights are derived by normalizing the CV using Equation 11.
After correction, it should satisfy: 0.01≤ ≤0.15 and
. The correction procedure is as follows: weights below 0.01 are set to 0.01, those above 0.15 are set to 0.15, and the resulting vector is subsequently re-normalized to mitigate the influence of extreme values in small samples.
GA-optimized comprehensive weight
The comprehensive weight , needs to be close to both the subjective weight of AHP and the objective weight of CV simultaneously. The objective function is defined as Equation 12:
Where: is the Manhattan distance between the comprehensive weight and the AHP weight (subjective deviation).
is the Manhattan distance between the comprehensive weight and the CV weight (subjective deviation).
,
is the weighting coefficient (since the objective weighting only took values from three location samples, subjective weighting is emphasized here).
The constraint conditions:
In the GA optimization process, real-number coding is employed. The chromosome represents a 25-dimensional weight vector . The initial population consists of 150 randomly generated weight vectors (
) that satisfy the prescribed constraints and are subsequently normalized.
The fitness function, as defined in Equation 13, is constructed such that a lower objective function value corresponds to higher fitness.
Genetic operations:
Selection: Tournament selection (randomly select 3 individuals and retain the one with the highest fitness);
Crossover: Simulated binary crossover (SBX), crossover probability 0.9, distribution index η = 20;
Mutation: Polynomial mutation, mutation probability 0.03, distribution index η = 20;
Termination condition: Iterate until the objective function converges or remains stable without significant change for 30 consecutive generations (change < 1 × 10−6).
Optimization result: The comprehensive weight obtained after GA optimization should satisfy:
Through iterative optimization, the final comprehensive weight retains the subjective rationality of AHP, absorbs the objective data characteristics of CV, and has a balanced weight distribution (standard deviation < 0.05).
The consistency test was performed using the Spearman rank correlation coefficient, as defined in Equation 9, to calculate the correlation between the comprehensive weight and the AHP weight.
Where, denotes the rank difference of the i-th indicator under the two weighting schemes, where n = 25. A Spearman correlation coefficient
> 0.6 indicates acceptable consistency between the two sets of weights. For the stability test, the comprehensive weight is slightly perturbed by ±5%, and the resulting change rate of the evaluation outcome is calculated using Equation 15.
If <10%, it indicates that the stability of the weight is acceptable.
Following the aforementioned steps, Python was utilized to implement matrix operations via NumPy and statistical tests via SciPy, enabling efficient execution of GA operations such as crossover and mutation, as well as statistical analyses including Spearman’s rank correlation coefficient. Calculation curves were generated using Matplotlib Fig 18–20. The weights for production areas, sales areas, and distribution centers were optimized through GA, yielding updated objective and comprehensive weights in Table 17. Iterative computations across all market types satisfied the constraint conditions and demonstrated satisfactory convergence, as show in Table 18.
Caption credit: This figure was created by the authors using PyCharm 2025.1.1.1.
Caption credit: This figure was created by the authors using PyCharm 2025.1.1.1.
Caption credit: This figure was created by the authors using PyCharm 2025.1.1.1.
Calculation of location selection results
Using the comprehensive weights optimized by the GA, the comprehensive scores for market location selection were recalculated according to Formula (16). The results, presented in Table 19, are consistent with the location selection recommendations obtained from the traditional calculation method shown in Table 10. Site A1 ranks first, followed by G1 and C1. This further confirms the accuracy and reliability of the location selection process under the established evaluation criteria.
Discussion
Discussion on criteria and weights.
Based on the analysis of comprehensive weights presented in Table 9 as show in (Fig 21), among the 16 secondary criteria, the top three for origin markets and distribution markets are L1, L2, and T5 (with F1 basic and T5 at an equivalent level), accounting for 39.40% and 38.82%, respectively. For destination markets, the leading criteria are L2, L1, and T5 (again with F1 basic and T5 at an equivalent level), collectively contributing 37.37%. This variation is primarily attributed to destination markets’ greater reliance on customer bases and population resources located in central urban areas. The subsequent three criteria—F1, P1, and F2—account for 29.17%, 29.45%, and 30.15%, respectively. It is evident that all market types exhibit substantial weighting toward factors such as connectivity between production and sales regions, proximity to the central urban area, and distance from airports. Additionally, sensitivity is observed regarding distance from similar markets, planned industrial concentration zones, and hazardous or polluted sources. The combined weight of the top six criteria exceeds 65% across all market categories.
Caption credit: This figure was created by the authors using OriginPro 2024b.
In previous research, Uyanık et al. reviewed 35 studies on logistics facility location [29], and the most frequently mentioned location criteria were Cost (land, labor, investment, facility, operation, transportation, information), followed by Natural Resources (water, power and electric supply, weather, etc.), and then Proximity to railroad, highways, airports, and labor force. This study also provides a review of the selection and classification of location criteria in Table 1. In comparison with earlier findings, under the emerging context of new circulation patterns, cost considerations between production and sales locations remain the most influential factor. However, advances in engineering, environmental adaptation technologies, transportation infrastructure, and related fields have gradually reduced the significance of natural resource factors. Proximity to transportation networks and labor resources continues to play a critical role. Notably, with improvements in circulation efficiency and quality, the expanding reach of distribution networks, and the development of the low-altitude economy, the importance of proximity to airports has substantially increased.
Following a comprehensive evaluation of the candidate schemes, this study conducted perturbation analysis on selected evaluation criteria by varying average vehicle speed and commuting radius duration. Simultaneously, the comprehensive weights were optimized using a GA. Notably, none of these adjustments altered the initial ranking of location selection advantages and disadvantages. This outcome indicates, on one hand, that the selection criteria and their corresponding weight assignments in this study are relatively stable and representative. On the other hand, it reflects the intercorrelation among certain criteria and the concentration of their weights. Furthermore, these findings indirectly suggest that, within a specific regional context, there exist relatively favorable zones for locating large-scale agricultural product wholesale market logistics facilities—highlighting both theoretical significance and practical research value.
Practical verification of location selection results
Research on logistics centers and logistics markets facilitates scientific decision-making in facility location. Studies on location selection models, influencing criteria, and methodologies have yielded diverse outcomes. However, given the evolving trends of modern circulation, there remains a lack of sufficient research on aligning the location selection of large-scale APWM with the demands of modern circulation. This paper examines the novel characteristics associated with the transition from traditional to modern circulation, develops a quantitative location selection evaluation system tailored for modern circulation, thereby contributing new strategies and approaches for constructing an efficient modern circulation system for agricultural products. Through case studies, the paper draws conclusions on optimal location selection. Further analysis of the location selection conclusions for the proposed PWM, DWM, and CWM in the case city, in conjunction with the layout of territorial and spatial planning and industrial planning, indicates that:
As a state-level and backbone PWM for agricultural products, the A1 location is strategically situated at the intersection of the continuous development zones of the two dominant producing areas in Qingdao City. It maintains strong connectivity with the characteristic agricultural product development zone, the coastal marine pasture development zone, and the fisherman’s wharf. Emphasis is placed on establishing close linkages with specialized producing area wholesale markets. Positioned at the core distribution hub of eight key areas, it encompasses the existing large-scale agricultural product wholesale market, the “14th Five-Year Plan” agricultural product processing park, the food processing enterprise cluster zone, the municipal demonstration professional cooperative society, and the family farm cluster zone. This strategic positioning facilitates the formation of four linkage corridors that connect various market entities within the city. As can be seen from (Fig 22a).
(a) The PWM circulation system under the newly selected location A1; (b) The DWM and CWM circulation system under the newly selected location A1. Caption credit: This figure was created by the authors using Adobe Illustrator. The base map is sourced from the “Territorial Spatial Planning of Qingdao City (2021-2035)” (Public Display Version). Available from: http://www.jimo.gov.cn/zwgk/bmxxgk/zrzyj/zdgk/ghjh/kjgh/202110/t20211016_3637933.shtml.
As the location for a DWM or CWM: within the consumption network, it is positioned closest to the primary urban area. It is centrally located relative to the city’s farmers’ markets, community markets, and the main supermarket POI clustering zones. The one-hour supply isochrone established by this location selection essentially covers 81.36% of the total urban development boundary area. Proximity to key agricultural product production areas in the northwest, as well as aggregation zones such as Jiaozhou, Pingdu, and Laixi, facilitates efficient collection and supply of agricultural products throughout the city. Optimal public transport accessibility supports the establishment of an integrated wholesale market that combines travel, shopping, research, learning, and entertainment functions. As can be seen from (Fig 22b).
In terms of logistics transportation, the location is strategically located near the Jiaodong International Airport and the Jiaozhou “Belt and Road Initiative” multimodal international freight railway land-port. Situated in the relatively central area of the intercity expressway entrance and exit network’s radiation zone, it can establish a logistics circulation corridor system that connects domestic and international destinations, extends throughout the country, reaches the Shandong Peninsula, and covers the entire city.
Expert discussion on the location selection results
To validate the reliability of the location selection outcomes derived from this study, the final location selection recommendations were submitted for review to a temporarily established expert review committee. This committee comprised specialists from the fields of market development, urban planning, logistics management, integrated transportation systems, and marketing, as well as representatives from key commercial stakeholders. The principal feedback and evaluations are summarized in Table 20.
Social impact discussion
Accessibility and Equity: Under the new circulation system, large-scale agricultural product wholesale markets are closely linked to major suppliers and centralized purchasers, while also connecting diverse individuals through emerging transaction platforms such as internet-based systems and mobile applications. First, the protection of disadvantaged groups is comprehensively addressed in this study through indicators including public transportation accessibility, a 15-minute convenient delivery network, and spatial layouts compatible with e-commerce operations. In the event of major natural disasters or public health emergencies, the scope of vulnerable populations may expand to encompass the majority of urban residents. Consequently, the accessibility of emergency support measures plays a significant role in the location selection rationale of this research. Simultaneously, a 15-minute driving service radius is established as a key inflexible criterion for ensuring equitable distribution of market resources within urban areas.
Urban-Suburban and Urban-Rural Integration: The new circulation framework connects a broader spectrum of logistical resources. With continuous advancements in logistics information technology, infrastructure, and transportation systems, resource allocation between urban and suburban areas, as well as between urban and rural regions, has become increasingly balanced. The location selection outcomes of this study are all situated in transitional zones where urban and rural areas intersect, incorporating weighted considerations related to agricultural product origins, destinations, and distribution centers. This balance extends to the formation of an integrated urban circulation network. By comprehensively evaluating various circulation elements across the entire metropolitan area, a balanced circulation structure is established—prioritizing accessibility under differentiated weighting schemes—laying the foundational conditions for realizing functional differentiation, hierarchical classification, and systemic coordination of the citywide agricultural product wholesale market system in the future.
Harmonization with Community Living Environments: Large-scale agricultural product wholesale markets generate substantial freight demand. To support daytime trading activities, significant volumes of goods are typically unloaded, stored, and transported during nighttime hours, potentially causing adverse effects such as noise pollution, traffic congestion, and waste accumulation in residential neighborhoods. In this study, “urban compatibility” is adopted as a critical evaluation criterion to prevent the introduction of such facilities from disrupting the integrity of urban living environments. Additionally, the potential for landscape resource development is incorporated as an assessment indicator, encouraging these markets to evolve into high-quality urban functional spaces that support leisure, recreation, and positive environmental experiences for local residents.
The reference value of location selection results
Globally, in the post-harvest stages of food production—including harvesting, storage, and transportation—the rate of food loss in developing countries is significantly higher than in developed countries, exceeding by approximately 40% to 67%. This indicates a negative correlation between food loss rates and the level of agricultural technology and infrastructure development [73].
In addition, in developing countries, there is a compelling need for a robust supply chain governance strategy to regulate the supply chain actors involved in the logistics and distribution system [74]. According to the World Bank’s comparison of the LPI of about 140 countries around the world, Over the past decade, high-income countries have occupied the top positions in the LPI rankings.
China’s logistics system has undergone significant improvement in parallel with its economic growth and infrastructure development. As an integral component of this progress, agricultural product logistics has also garnered increased attention and investment. This study analyzes the LPI data [75] for various countries as provided by the World Bank, presenting results for seven developing nations—China, Malaysia, Brazil, India, and others. The findings reveal that between 2014 and 2022, these countries consistently achieved LPI scores above the average of both the top 10 and bottom 10 performers in their respective years, indicating relatively strong logistics performance overall (Fig 23). Given the common trajectories in economic and logistics sector development among developing countries, this analysis may serve as a valuable reference for those currently undergoing rapid modernization of their agricultural product logistics systems.
IS, International Shipment; LCQ, Logistics Competence and Quality; TT, Tracking and Tracing. Caption credit: This figure was created by the authors using OriginPro 2024b. The data is sourced from the World Bank’s evaluation reports on the Logistics Performance Index (LPI) of the world’s economies from 2014 to 2022. https://lpi.worldbank.org/international/global.
Conclusion and future work
The diversity of relevant criteria involved in the location selection for large-scale APWM, as well as the complexity of their relationship with urban and rural areas, constitute the core challenges of location selection. In response to the new demands of modern circulation.
This study examines five major transitions from traditional to modern circulation, summarize existing research findings, and construct an integrated evaluation system for the location selection of large-scale APWM suited for modern circulation. Achieve the integration of territorial and spatial planning, development planning, and urban big data resources in the location selection process. Simultaneously, this approach highlights the integration of a three-dimensional logistics network encompassing land, sea, and air dimensions. A hybrid method combining GIS technology with the AHP and entropy weight methods is proposed, and the comprehensive weights are optimized using GA.
This method, through application in actual case cities, progressively enables regional screening, scope delineation, data analysis, weight optimization, and comprehensive evaluation. It ultimately recommends three alternative plans and conducts a quantitative ranking to identify the optimal solution. Although the ranking and results remained largely unchanged after perturbation and GA optimization, the findings demonstrate that the proposed evaluation system and methodology are effective and stable. Furthermore, it provides scientific support for the location selection of major project planning. This research also serves as a valuable reference and provides guidance for the subsequent development of specialized and detailed plans for territorial space.
In the era of modern circulation, constructing a rational and comprehensive evaluation system and indicator framework to establish a modern agricultural products circulation network space system requires further in-depth research.
In the future, when developing agricultural product wholesale markets or other similar logistics infrastructure systems under the new circulation framework, greater emphasis should be placed on the role of facility location evaluation indicators in promoting social and environmental sustainability. At the same time, the impact of emerging technologies—such as drones, automated cold chain logistics, and e-commerce—must be carefully considered, along with their performance across diverse application scenarios influenced by extreme weather events, diurnal variations, and emergencies such as pandemics. The siting of critical infrastructure essential to maintaining the basic living standards of urban and rural populations should be integrated into regional or municipal emergency preparedness and recovery plans.
Supporting information
S1 Data. Publicly available datasets for GIS analysis.
https://doi.org/10.1371/journal.pone.0345727.s002
(ZIP)
S1 Fig. The Third Phase Construction Plan of Qingdao Urban Rail Transit (2021–2026) (Draft for Public Display).
https://doi.org/10.1371/journal.pone.0345727.s003
(JPG)
Acknowledgments
The authors would like to acknowledge Qingdao Conson Development (Group) CO., LTD for providing valuable information on market development; as well as Qingdao Urban Planning & Design Research Institute for the consultation on the relevant planning. We gratefully acknowledge Professor Tang Yuqing for his conceptualization and suggestions.
References
- 1.
Li Z. Public welfare of agricultural products wholesale markets: Evaluation and path design. China Agricultural University. 2014.
- 2.
China Agricultural Wholesale Market Association. Farmers’ Market Development Survey Report 2021 (In Chinese). Beijing: China Agricultural Wholesale Market Association. 2022. https://www.cawa.org.cn/index.php?a=show&id=13788_1673_44
- 3. Levi R, Rajan M, Singhvi S, Zheng Y. The impact of unifying agricultural wholesale markets on prices and farmers’ profitability. Proc Natl Acad Sci U S A. 2020;117(5):2366–71. pmid:31964822
- 4. Ruan J, Cai Q, Jin S. Impact of COVID-19 and Nationwide Lockdowns on Vegetable Prices: Evidence from Wholesale Markets in China. Am J Agric Econ. 2021;103(5):1574–94. pmid:33821009
- 5.
All China Federation of Supply and Marketing Cooperatives. Notice of the General Office of the All China Federation of Supply and Marketing Cooperatives for Public Consultation on the Standards for the Construction of Wholesale Markets for Agricultural Products (Revised) (Draft for Comment) (In Chinese). https://www.chinacoop.gov.cn/news.html?aid=855264. 2015. Accessed 2024 December 7.
- 6.
Institute of Food Science and Technology CAAS. Guidance on risk prevention and control of vegetable collection, storage and transportation in China. 2022. http://www.jgj.moa.gov.cn/fxpg/202202/t20220225_6389723.htm
- 7.
Ministry of Agriculture and Rural Affairs of the People’s Republic of China. Circular of the Ministry of Agriculture and Rural Affairs on the issuance of the “14th Five-Year Plan” for the development of the national agricultural products market system. 2022. http://www.moa.gov.cn/nybgb/2022/202203/202204/t20220401_6395165.htm
- 8. Fisher-Holloway B, Mokhele M. The deployment of theory in logistics research: A systematic literature review. J transp supply chain manag. 2022;16.
- 9. Borth K, Summers R, Collins D. Conceptualizing location efficiency: a narrative review and scale-based consolidated definition. J Plan Educ Res. 2024;44(3):1595–606. pmid:WOS:001294442100023
- 10. Horton D, Murrell J, Skipper D, Speakman E, Logan T. A scalable optimization approach for equitable facility location: methodology and transportation applications. Transp Res Pt B-Methodol. 2025;201:28. pmid:WOS:001587187800001
- 11. Sahin M. Ensemble decision making for logistics center location. Environ Dev Sustain. 2024;26(12):30527–61. pmid:WOS:001080599700002
- 12. Huang Y, Huang XN, Chen YW, Zhu J. Research on spatial pattern evolution and driving mechanism of county express delivery network in Hunan Province. Economic Geography. 2025;45(04):56–66.
- 13. Chen YT, Cai XN, Dai JY. Evolution of the circulation mode of fresh agriculture products: from traditional farmers market to fresh supermarket. China Business and Market. 2013;27(03):19–23.
- 14. Hong T. Report on 70 years of development of agricultural products circulation in China. Journal of Commercial Economics. 2019;21:190–3.
- 15. Cadilhon J-J, Moustier P, Poole ND, Tam PTG, Fearne AP. Traditional vs. Modern Food Systems? Insights from Vegetable Supply Chains to Ho Chi Minh City (Vietnam). Development Policy Review. 2006;24(1):31–49.
- 16. Oparebea Boateng A, Kwasi Bannor R, Bold E, Helena O-K. A systematic review of the supply of agriproducts to supermarkets in emerging markets of Africa and Asia. Cogent Food & Agriculture. 2023;9(1).
- 17. Xu JY. Research on the high quality development of modern circulation industry empowered by new quality productivity. Theoretical Exploration. 2024;(06):13–9.
- 18. Danyluk M. Capital’s logistical fix: Accumulation, globalization, and the survival of capitalism. Environ Plan D. 2017;36(4):630–47.
- 19. Cichosz M. Digitalization and competitiveness in the logistics service industry. E-mentor. 2018;77(5):73–82.
- 20. Cichosz M, Wallenburg CM, Knemeyer AM. Digital transformation at logistics service providers: barriers, success factors and leading practices. IJLM. 2020;31(2):209–38.
- 21. Hu QB. Significance and practical requirements for the construction of a modern circulation system under the new development paradigm. Logistics Research. 2025;(1):81–6.
- 22. Chen Wl. The Form Change and Theoretical Reconstruction of Modern Circulation. Frontiers. 2022;(07):78–101. Chinese.
- 23. Dindar S. A systematic review of urban regeneration’s impact on sustainable transport: traffic dynamics, policy responses, and environmental implications. Sustain Dev. 2025;28. pmid:WOS:001511492800001
- 24.
Xiao LX. Research on the location of cold chain logistics distribution center for fresh agricultural products. Dalian Jiaotong University. 2023.
- 25. Awasthi A, Chauhan SS, Goyal SK. A multi-criteria decision making approach for location planning for urban distribution centers under uncertainty. Mathematical and Computer Modelling. 2011;53(1–2):98–109.
- 26. Agrebi M, Abed M. Decision-making from multiple uncertain experts: case of distribution center location selection. Soft Comput. 2020;25(6):4525–44.
- 27. Dey B, Bairagi B, Sarkar B, Sanyal SK. Warehouse location selection by fuzzy multi-criteria decision making methodologies based on subjective and objective criteria. International Journal of Management Science and Engineering Management. 2015;11(4):262–78.
- 28. Yazdani M, Chatterjee P, Pamucar D, Chakraborty S. Development of an integrated decision making model for location selection of logistics centers in the Spanish autonomous communities. Expert Systems with Applications. 2020;148:113208.
- 29. Uyanık C, Tuzkaya G, Oğuztimur S. A literature survey on logistics centers’ location selection problem. Sigma Journal of Engineering and Natural Sciences. 2018;36(1):141–60.
- 30. Zhang YY. Site selection model of agricultural products wholesale market based on GIS and entropy weight TOPSIS evaluation methods--taking the proposed municipal agricultural products wholesale market in Qingdao as an example. Building Technology. 2024;8(1):26–30.
- 31. Onstein ATC, Tavasszy LA, van Damme DA. Factors determining distribution structure decisions in logistics: a literature review and research agenda. Transport Reviews. 2018;39(2):243–60.
- 32. Muerza V, Milenkovic M, Larrodé E, Bojovic N. Selection of an international distribution center location: A comparison between stand-alone ANP and DEMATEL-ANP applications. Research in Transportation Business & Management. 2024;56:101135.
- 33. Ocampo LA, Himang CM, Kumar A, Brezocnik M. A novel multiple criteria decision-making approach based on fuzzy DEMATEL, fuzzy ANP and fuzzy AHP for mapping collection and distribution centers in reverse logistics. Adv produc engineer manag. 2019;14(3):297–322.
- 34. Özmen M, Aydoğan EK. Robust multi-criteria decision making methodology for real life logistics center location problem. Artif Intell Rev. 2019;53(1):725–51.
- 35. Pham TY, Ma HM, Yeo GT. Application of Fuzzy Delphi TOPSIS to Locate Logistics Centers in Vietnam: The Logisticians’ Perspective. The Asian Journal of Shipping and Logistics. 2017;33(4):211–9.
- 36.
Yu Y. A study on site selection decision of agricultural products wholesale market in Wuhan city circle. Huazhong Agricultural University. 2011.
- 37. Song ZL, Huang Y, Kong MJ, Bai XY, Li EL, Gu CY. Study on site selection of fresh agricultural product logistics distribution center based on TOPSIS and GRA: An empirical study. Logistics Technology. 2019;38(12):49–53.
- 38.
Li N. A study on the location of agricultural products logistics center. Beijing Jiaotong University. 2009.
- 39. Kieu PT, Nguyen VT, Nguyen VT, Ho TP. A spherical fuzzy analytic hierarchy process (SF-AHP) and combined compromise solution (CoCoSo) algorithm in distribution center location selection: A case study in agricultural supply chain. Axioms. 2021;10(2). pmid:WOS:000665133600001
- 40. Raimbekov Z, Syzdykbayeva B, Rakhmetulina A, Rakhmetulina Z, Abylaikhanova T, Doltes L, et al. Designing and improving the efficiency of placement of logistics infrastructure for storage, distribution and trade of agri-food products. JIEM. 2024;17(3):777.
- 41. Giuffrida N, Calleo Y, Di Zio S, Pilla F, Ottomanelli M. Spatial participatory planning for urban logistics: A GIS-enhanced Real-Time Spatial Delphi approach. Research in Transportation Economics. 2024;108:101488.
- 42. Charters-Gabanek K, Raimbault N, Hall P. Logistics and urban planning: a review of literature. J Plan Educ Res. 2025;45(3):663–73. pmid:WOS:001207145300001
- 43. Kin B, Rai H, Dablanc L, Quak H. Integrating logistics into urban planning: best practices from Paris and Rotterdam. Eur Plan Stud. 2024;32(1):24–44. pmid:WOS:001038649500001
- 44. Wang Y, Li Y, Huang Y, Gong D. Analyzing the impacts of logistics suburbanization on logistics service accessibility: Accessibility modeling approach for urban freight. Transport Policy. 2023;138:25–44.
- 45. Chen J, Lee J, Lee H. A study on the site selection of urban logistics centers utilizing public infrastructure. Sustainability. 2025;17(15):20. pmid:WOS:001549805100001
- 46. Abdulrazaq MA, Fan W (David). A priority based multi-level heterogeneity modelling framework for vulnerable road users. Transportmetrica A: Transport Science. 2025;:1–34.
- 47. Wang XQ. The strategic focus of modern logistics system and the policy choice under the background of new development pattern. China Business and Market. 2020;34(11):18–32.
- 48. Hu WJ, Wang LY. Optimization and change of “modern circulation” industry chain organization mode. Journal of Commercial Economics. 2019;(22):5–8.
- 49. Sun QJ. Exploring and doing research on the frame composition of China’s modern circulation system. China Business and Market. 2011;25(10):12–6.
- 50. Ding J. Accelerating the construction of efficient modern circulation system. Economic Daily. 2020.
- 51.
Department of Market System Construction M of C. National Agricultural Products Market System Development Plan (In Chinese). Beijing: Department of Market System Construction. 2015. http://scjss.mofcom.gov.cn/zcfb/art/2015/art_e76f5b3b1efb424b95db5ddb01e8357f.html
- 52. Fan LH, Wang PF. The spatial and temporal evolution of China’s wholesale market for agricultural products and its commodification effect. Economic Geography. 2019;39(07):175–83.
- 53. Sakai T, Santo K, Tanaka S, Hyodo T. Locations of logistics facilities for e-commerce: a case of the Tokyo Metropolitan Area. Research in Transportation Business & Management. 2024;56:101174.
- 54. Gingerich K, Maoh H. The role of airport proximity on warehouse location and associated truck trips: Evidence from Toronto, Ontario. Journal of Transport Geography. 2019;74:97–109.
- 55. De Silva CK, Sano K, Hatoyama K. Exploring the relationship between urban form and spatial organisation of Amazon fulfilment facilities in the United Kingdom and Japan. Transportation Research Procedia. 2020;46:149–56.
- 56. Gibbons S, Wu W. Airports, access and local economic performance: evidence from China. Journal of Economic Geography. 2019;20(4):903–37.
- 57.
Chang’an University, SINOIOV. Annual report on big data analysis of highway freight transport in China. Beijing: SINOIOV. 2024. https://www.sinoiov.com/news/official/436.html
- 58.
General Office of the Chongqing Municipal People’s Government. Opinions of the General Office of the Chongqing Municipal People’s Government on promoting the scientific distribution and sustainable healthy development of agricultural products wholesale markets in the main urban area. Chongqing Municipal People’s Government. http://wap.cq.gov.cn/zwgk/zfxxgkml/szfwj/xzgfxwj/szfbgt/202111/t20211116_9980706.html. 2021. Accessed 2025 January 6.
- 59. Bohannon RW, Williams Andrews A. Normal walking speed: a descriptive meta-analysis. Physiotherapy. 2011;97(3):182–9. pmid:21820535
- 60.
Ministry of Housing and Urban-Rural Development of the People’s Republic of China. Announcement of the Ministry of Housing and Urban-Rural Development on the release of the national standard “Urban comprehensive transportation system planning standard. Ministry of Housing and Urban-Rural Development of the People’s Republic of China. https://www.mohurd.gov.cn/gongkai/zc/wjk/art/2019/art_17339_239844.html. 2019. Accessed 2025 January 9.
- 61.
Qingdao Market Supervision and Administration Bureau. Specification for construction and management of farmer’s market. Qingdao: Qingdao Market Supervision and Administration Bureau. 2022. http://sjw.qingdao.gov.cn/cxjsj15/cxjsj102/202210/P020221027596183578722.pdf
- 62. Álvarez P, Serrano-Hernandez A, Lerga I, Faulin J. Optimizing freight delivery routes: The time-distance dilemma. Transportation Research Part A: Policy and Practice. 2024;190:104283.
- 63.
U.S. Department of Transportation. Average Truck Speeds on Select Metropolitan Area Interstates. 2020. https://www.bts.gov/browse-statistical-products-and-data/freight-facts-and-figures/average-truck-speeds-select
- 64.
General Office of the People’s Government of Qingdao City. Notice of the General Office of the Qingdao Municipal People’s Government on Printing and Distributing the Urban Construction Plan for the Modern Circulation Strategic Fulcrum of Qingdao (In Chinese). http://www.qingdao.gov.cn/zwgk/zdgk/fgwj/zcwj/szfgw/gw_2024/qzbz/202412/t20241203_8650512.shtml. 2024. Accessed 2025 March 25.
- 65.
Ministry of Housing and Urban-Rural Development of the People’s Republic of China, Urban Transportation Infrastructure Monitoring and Governance Laboratory, China Academy of Urban Planning & Design, NavInfo, Baidu Map. Annual Report on Road Network Density & Traffic Operation in Major Chinese Cities. Beijing: Ministry of Housing and Urban-Rural Development of the People’s Republic of China. 2024. https://mp.weixin.qq.com/s/B14P4GxERmPl22W1NlnFgg
- 66.
National Development and Reform Commission of the People’s Republic of China. Approval on the Third Construction Plan of Urban Rail Transit in Qingdao City (2021-2026) (In Chinese). Beijing: National Development and Reform Commission of the People’s Republic of China. 2021. https://new.tzxm.gov.cn/zckd/gfxwj/202109/W020210910686231168725.pdf
- 67.
Qingdao Metro Group Limited. Qingdao Metro social responsibility report 2023 (In Chinese). Qingdao: Qingdao Metro Group Limited. 2024. http://www.qd-metro.com/news/view.php?id=7225#pagetop
- 68.
General Office of the People’s Government of Qingdao City. Opinions of the General Office of the People’s Government of Qingdao City on Further Promoting the Priority Development of Urban Public Transport (In Chinese). Qingdao: General Office of the People’s Government of Qingdao City. 2023. http://www.qingdao.gov.cn/zwgk/xxgk/bgt/gkml/gwfg/202308/t20230815_7412698.shtml
- 69. Hu L, Liu H, Ying C. Construction of hub location model based on genetic algorithm. Transp Res Record. 2025;2679(2):403–20. pmid:WOS:001279805100001
- 70. Liu J, Li Y, Li Y, Chen Z, Lian X, Zhang Y. Location optimization of emergency medical facilities for public health emergencies in megacities based on genetic algorithm. Eng Constr Archit Manag. 2023;30(8):3330–56. pmid:WOS:000818972900001
- 71. Yan L. Robust optimization of urban cold-chain logistics location-routing based on an improved genetic algorithm. Univ Politeh Buchar Sci Bull Ser C-Electr Eng Comput Sci. 2025;87(1):139–56. pmid:WOS:001445507200010
- 72. Ayodeji A, Mujeeb Abiola A. Defining accuracy benchmarks for freeway traffic simulations in support of highway operations and planning. World J Adv Res Rev. 2025;27(2):790–7.
- 73. Wang F. Promote green reduction of grain loss through scientific and technological innovation (In Chinese). Farmers’ Daily. https://ifst.caas.cn/xwzx/mtbd/285426.htm. 2021. Accessed 2025 October 10.
- 74. Perdana T, Tjahjono B, Kusnandar K, Sanjaya S, Wardhana D, Hermiatin FR. Fresh agricultural product logistics network governance: insights from small-holder farms in a developing country. International Journal of Logistics Research and Applications. 2022;26(12):1761–84.
- 75.
Word T, Bank. Logistics Performance Index (LPI). Washington: The Word Bank. 2023. https://lpi.worldbank.org/international/global