Evaluation of smart city logistics solutions with fuzzy MCDM methods

Abstract


Introduction
The majority of the population in the world and also majority of Turkey lives in cities. Therefore, cities are the most complex and significant impact on our daily life. Especially, logistics is the most essential area that effects city residents' life. Parallel to population growth, the use of private vehicles augments. And that reveals the need for the increment of the traffic, air pollution, transportation costs in cities and likewise the product prices. City logistics, which started to examine as a subdivision of logistics at the beginning of the 1990s, aims to plan and to manage the distribution and transportation [1].
The primary objectives of city logistics are efficiency, protection of the environment, reduction of traffic, security, and energy-saving. These aims are related to the three cornerstones of city logistics that are sustainability, mobility, and livability. Parallel to the rapid development of Internet and Information Technologies, logistics becomes one of the affected and developing sectors. "Smart city logistics solutions" are presented by Information Technologies to meet the increasing demand for logistics services more efficiently, safely, and environmentally [2]. Smart city logistics is a fundamental approach for the more efficient organization of physical and information logistics in international transportation chains. It means an optimized form for the transport of people and goods, which increases mobility, security, and the benefits of users and at the same time, reduces pollution, consumption, and bottlenecks [3]. Sustainability and livability, which are the cornerstones, are the concepts that provide the continuity of smart city logistics [4].
Logistics networks are exposed to uncertain environments. Therefore, the need for robustness, flexibility, and agility has become a focal point for future logistics system designs. Smart city logistics solutions have developed to reduce the harmful effects of increased transport in urban areas [5].
In this study, the evaluation of smart city logistics solutions that contain many components, is approached as a multi-criteria decision making (MCDM) problem. Considering the complex profile of this problem, this complexity is needed to be taken into account by experts for deciding on a suitable solution. However, it is challenging to decide on the most appropriate solution when information is in an uncertain nature.
This study aims to evaluate smart city logistics solutions with analytical methods. In this context, the smart city logistics solutions, which include customer and technological requirements in Istanbul determined by literature review and expert opinions, are modeled. This model is evaluated by using the House of Quality matrix of Quality Function Deployment (QFD) approach with fuzzy MCDM methods. Fuzzy Simple Additive Weighting (SAW) method is used to calculate the weights of customer requirements.
The SAW is a straightforward and widely used MCDM method. It is mainly used in one-dimensional decision-making problems. The weighted sum of the performance values of each alternative is obtained according to all criteria [6]. The House of Quality matrix is a tool that determines the customers' expectations and requirements . These customer requirements, which are determined by literature review and expert opinions, should have the importance degree for the evaluation of technological requirements. In this study, the importance of customer requirements is calculated with fuzzy SAW method. In addition, this method is supported by linguistic expressions. Therefore provides more realistic results with uncertain data and it provides a flexible environment for decision process [7].
The structure of the paper is as follows: The related studies about smart city logistics summarized in the next section. The third section presents the proposed model and methodology. The fourth section presents the application, and finally, conclusions have provided.

Smart city logistics
Smart city logistics is a fundamental approach to reach more efficient organization for physical and information logistics in international transportation chains. This approach provides economic, environmental, and social sustainable solutions by ensuring that information can be achieved quickly and efficiently [8].
The goals of smart city logistics are to provide multidimensional data exchange between passenger-vehicleinfrastructure-center, to increase traffic safety and mobility, to make use of roads in accordance with their capacity, to reduce energy loss by providing energy efficiency [9]. Smart city logistics applications can provide coordination between different modes of transportation to create ideal traffic conditions and increase the efficiency and speed of services related to freight movements. By optimizing travel times and reducing the risk of crashes and injuries, the performance of modern transport systems is improved [10].
Today, smart city logistics is a system based on advanced technologies in the regulation and management of transportation. These systems use real-time and up-to-date databases, and they serve to improve efficiency, safety, and service quality in transportation. There are many benefits of smart transportation with systems such as traffic tracking systems, advanced passenger information systems, pricing systems, advanced transportation management systems, and advanced public transportation systems [11].
There are several studies about smart city logistics in the literature. Some of these studies have shown in Table 1. According to Table 1, there are generally theoretical studies related to this subject, and the number of studies with the application area is small. In this study, this subject is proposed with an application area, Istanbul.

Proposed model and methodology
In this paper, the aim is to propose a methodology for determining and prioritizing the most crucial technological requirement for smart city logistics. In this context, the study proposes an integrated methodology.
A three-phase methodology in this study is as follows: Phase 1: Determination of customer and technological requirements in the House of Quality matrix for smart city logistics.
Phase 2: Determination of customer requirements' weights by fuzzy SAW method.
Phase 3: Evaluation of the relationship between customer requirements and technological requirements, prioritization of technological requirements.
The flow chart of this methodology is given in Fig. 1.

The house of quality matrix of the QFD
QFD is an important design technique that enables the transformation of customers' needs to product or service characteristics in all functional components, and it is a planning, development, and communication tool [38]. QFD is a strategic tool used to develop improved products and services responsive to customer needs. This approach aims to translate customer requirements into appropriate technical or technological requirements.
The House of Quality is the most frequently used matrix in the QFD process. The House of Quality matrix is conducted by a team of multidisciplinary experts to translate customer requirements from market research and to benchmark data into technical requirements which will be met by designing a new product or service [39]. This matrix is used to evaluate the technological requirements in the smart city logistics system by meeting customer needs. With this approach, customer requirements have determined, and it is aimed to increase customer satisfaction by meeting the needs of customers. At the same time, design optimization and efficiency will improve. It is a systematic algorithm that is used as the essential design tool of QFD. It organizes data and establishes relationships [40].
The House of Quality matrix consists of two main parts. The first part is the part of the customer requirements, and the second one is the part of the technological requirements.

Customer Requirements
In the smart city logistics context, the determined customer requirements are as follows:  Accessibility and travel time (CR1): Limited access in some cases increases travel quality. It is the restriction of individual vehicles on certain roads or areas. The limitation may be implemented on specific modes or periods. For example, borders to reduce the intensity in the city center during peak hours, the use of only environmentally friendly vehicles in some particular cases such as low emission zones, etc. Thus, there is a decrease in traffic intensity, an increase in the speed of public transportation, an improvement in the quality of the air, and an increase in pedestrian safety [41].  [12] to improve the smart city logistics and evaluate the required strategies with analytic methods -

2019
Korczak & Kijewska, [13] to present the use of the smart city logistics potential in the development of smart cities -

2018
Taniguchi & Thompson [14] to develop the current city logistics model -

2018
Stoller & Wan [15] to development of smart city logistics China 2018 Kaźmierczak et al. [16] to present the current state of smart city logistics project Poland 2018 Bachanek [17] to present practical methods for implementing the concept of management urban space -

2018
Gruler [18] to present simheuristics to support efficient and sustainable freight transportation in smart city logistics -2017 Eitzen et al. [19] to propose an urban goods movement for smart city logistics -2017 Shuai & Hong-Chun [20] to present a smart city logistics concept China 2017 Bektaş et al. [21] to present from urban freight to smart city logistics networks - Nowicka [28] to present smart city logistics on the cloud computing model -2014 Malecki et al. [29] to present influence of Intelligent Transportation Systems (ITS) Szczecin

2014
Taniguchi [30] to present city logistics for sustainable and livable cities -2014 Thompson & Hassall [31] to present high productivity freight vehicles in urban areas -

2014
Zidi et al. [32] to propose a solution for congestion with an intelligent support system -2014 Cagliano et al. [33] to present e-grocery by developing an ITS -2014 Lu [34] to present an innovative solution -2011 Tesch et al. [35] to present an innovative approach for road transportation -2010 Oliveira et al. [36] to present the intelligent delivery points Brazil

2007
Ambrosini & Routhier [37] to compare the objectives, methods, and results of urban freight transport USA, Canada, Australia, Japan, UK, Germany, Switzerland, The Netherlands, France  Road pricing (CR2): Road pricing is about the size of the cities and transportation capacities. Price areas are determined, and tariffs have created accordingly. In the peak periods, the tariffs may change according to the hours of the day. Top of the pricing, systems is electronic payment systems; bottleneck-based pricing, free express lines, vehicle mileage-based payment systems [42].
 Lane management (CR3): Lane management is the flexible use of lanes in certain situations. Specific lanes reserved for public transport, temporary closure of roads in cases of accident, maintenance work and construction measures and the addition of extra lanes during peak hours are the lane management works [4].
 Public transportation priority (CR4): The public transport priority is the setting of traffic control sets to minimize the stopping times of public transport. The timing of the traffic lights should be arranged accordingly. The public transport system needs to be made more attractive than individual vehicle use. For this reason, it is necessary to shorten long waiting times, to provide uninterrupted and safe transportation, and at the same time, the fees should be reasonable for the users [9].
 Park guidance (CR5): The drivers need to be directed during parking in the parking place. Park guidance presents dynamic information to drivers for parking in controlled areas. It is designed to help drivers navigate to parking garages and look for empty parking spaces. Thus, traffic volumes reduced. The accessibility of the city center increased. The use of parking capacity is optimized [41]. is possible to plan the travel based on the individual or public transport options [44].
 Network robustness and road safety (CR7): Robust and reliable roads, bridges, and other infrastructure elements are essential for transport quality. Controls of these networks should frequently be fulfilled, and if a problem is encountered, it should be intervened immediately. In Turkey, CCTV cameras control public vehicles and stops. Also, elderly or disabled citizens can safely use the transportation system with the Accessible Pedestrian Button System developed by İSBAK [9].

Technological Requirements
In the smart city logistics context, the assigned technological requirements are as follows:  Demand and access management (TR1): Demand and access management usually implemented with the use of additional implementation strategies and variable message markings. It focuses on reducing the intensity of congestion and air pollution in large cities [41]. It determines the density according to the demand of the users and performs the pricing accordingly. It facilitates access to transportation means and services for those with restricted mobility.
 Full adaptive traffic management system (TR2): Full Adaptive Traffic Management System is a working system in which the parameters have optimized for minimizing average vehicle delay times and average stop numbers. It accelerates traffic flow by intervening in real-time on blocked roads and reduces delay times. It creates ideal signal times for the system. Thus, it reduces travel time and emissions on the road network [45].
 Security and emergency systems (TR3): A system that detects and prevents emergency intervention events such as traffic accidents [43]. Security systems consisting of cruise control systems and anti-lock braking systems have been developed. New generation systems such as emergency brake assist, emergency brake-force distribution, electronic stability control, and advanced speed control systems have been designed [9].  Variable message system (TR4): Variable Message Systems graphics-based fonts, shapes, and images can be displayed using LEDs for traffic purposes. Messages are used to inform drivers about traffic intensity, traffic accidents, weather, and road conditions and to control traffic flow. Thus, road traffic capacity could more efficiently be used to reducing local traffic densities [46].
 Intelligent public transport systems (TR5): Public transport is one of the essential parts of city logistics. Improvement of public transportation systems with the help of developing technologies also improves transportation conditions. The system that enables the most convenient use of the public transportation needs that arise with the increasing population [43]. Passenger information systems and electronic payment systems are the most common methods used in public transport. The purpose of the passenger information system is to prevent the inefficient use of time and energy and encourage public transportation. Advanced passenger information systems include smart stops and contactless smart cards [4]. Similar applications of intelligent systems used in public transportation are also available in private vehicles.

 Road and meteorological observation sensors (TR6):
These include sensors that support the drivers for safe travel and road and weather sensors. Road sensors are sensors that collect data to detect traffic flow information in the city. Weather sensors are powerful and versatile sensors that measure atmospheric conditions, monitor ground temperatures, and measure moisture and heat [44].
 Lane control systems (TR7): Lane control system includes an intelligent vehicle system and automatic road system. This system aims to reduce traffic accidents and ensure the safety of drivers. The system follows the lane markings on the road and gives the warning to return to the lane when the vehicle goes out of the lane [9].
 Parking management systems (TR8): An intelligent system that provides the utilization of parking spaces more  [43].
 An electronic detection system (TR9): It is a system, which the vehicles that violate the rules have detected with the sensors, the plates have recorded, and the drivers are subject to penalty. Using cloud technology, it has 12 different violation detection systems [41]. It has been developed in order to detect the parking lane, safety lane, pause violations that affect the traffic flow in the city logistics negatively and to prevent traffic problems caused by these violations, to prevent possible accidents by performing security control of the region with moving cameras, and to provide safety of lives and properties for pedestrians and vehicles [41,44].

Fuzzy SAW method
The SAW method, also known as the weighted sum method, is the most widely used MCDM method [6]. The basic principle of SAW is to obtain a weighted sum of the performance ratings of each alternative. An evaluation score is calculated for each alternative. The advantage of this method is that there is a proportional linear transformation of raw data; this means that the relative order of the sizes in the standardized points remains the same [47].
Chou et al. [48] proposed the fuzzy SAW method to solve problems under fuzzy environment. The steps of the fuzzy SAW method are as follows: Step 1. Decision makers (DM) evaluate criteria using linguistic terms in Table 2.
Step 2. Let Dt = {d1,d2,...,dk } be a committee of k DMs, Ai = {a1,a2,...,aI } be a discrete set with I member alternatives, Cj = {c1,c2,...,cJ } be a set consisting of the decision criteria, be the degree of importance of each DM, where 0 ≤ ≤ 1, t = 1,2,...,k, and , ̃be the fuzzy weight of the DMs. The degree of importance is computed as: where ( ) gives the defuzzified value of the fuzzy weight by using the signed distance.

Application
In this part of the study, an application is presented to demonstrate the applicability of the proposed research methodology. Expert opinion is needed in the application of the proposed methodology to identify smart city logistics solutions.
In this context, opinions were taken from three experts, and the steps of the proposed methodology were applied with the opinions of these DMs. The first expert, DM1, has logistics sector experience exceeding ten years. DM2 has experience for the smart systems projects in the logistics sector. DM3 is working on supply chain and logistics management and smart systems for city transportation in the academy. All three experts are sufficiently knowledgeable and experienced in the field of smart city logistics.
Step 1: Customer requirements of smart city logistics in House of Quality matrix are determined with a detailed literature survey and expert opinions.
Step 2: Technological requirements of smart city logistics in House of Quality components are determined with a detailed literature survey and expert opinions.
Step 3: DMs evaluated customer requirements using linguistic terms in Table 2 to calculate weights. These evaluations with linguistic expressions are shown in Table 3. Step 4: The equations (1) -(4) are applied and the weights of customer requirements are calculated as shown in Table 4. Step 5: DMs made a collective decision and formed the relationship matrix. For each actor group such as customer requirements, technological requirements, comparison matrices between them, weights and relationship matrix are placed in the House of Quality as Table 5.
Step 6: The importance degree of each technological requirement by multiplying the weight of the customer requirement and the relationship matrix values are found as follows: Weight of TR1 = [(0.168×9) + (0.084×9)]/16.112 = 0.141 The final importance degrees are shown in Table 5. As a result of the evaluation, the priority order of the technological requirements is as follows: security and emergency systems (TR3), full adaptive traffic management system (TR2), demand and access management (TR1), intelligent public transport systems (TR5), variable message system (TR4), communication systems (TR10), lane control systems (TR7), road and meteorological observation sensors (TR6), electronic detection system (TR9) and parking guidance systems (TR8).
According to this result, the first issue to focus is on smart city logistics solutions is security and emergency systems. It is the most critical solution because it has the highest evaluation of the relationship between customer requirements and technological requirements. At all times, human safety is a critical element. Thanks to smart systems in the case of emergency, fast and direct interventions protect passenger and driver's health. With the developed brake and speed control systems, a more secure logistics system will establish.

Conclusion
Countries with high economic levels make great strides in smart city logistics area and use the technology at the highest level. The critical need for smart city logistics may be understood if the material and spiritual losses that are caused by traffic accidents and congestions in our country are taken into account. It is vital to integrate the right technology to implement the required system.
This study aims to identify and evaluate smart city logistics solutions with fuzzy analytic methods. Fuzzy analytic methods are enables to decide where to start the development of a more sustainable, mobile, and livable city. In this reason, an integrated methodology is proposed. The components of the House of Quality determined by the literature survey and the expert opinions are prioritized by the proposed methodology. In this combined work, the priority ranking is made for all technologies without focusing on only one smart technology. According to the results obtained, safety and emergency systems have ranked first in importance.
With these systems, events that require any kind of emergency response are identified, the center concerned are informed, the incident is intervened, and the destruction caused by the incident is managed. Preventing the loss of life and minimizing material damage depends on the effective management of this process. In general, one of the most effective methods of event detection is camera monitoring and automatic alerting [41,46].
In future studies, the number of customer and technological requirements can be increased, and the problem can be solved using aggregation operators for group decision making to aggregate DMs' evaluations. Another perspective can be to consider uncertainty using an advanced fuzzy approach such as hesitant fuzzy linguistic term sets, intuitionistic fuzzy sets, etc.