Gıda ürünlerinin şehir-içi dağıtımının optimizasyonu için sürdürülebilir bir yaklaşım

Abstract

Gıda sektöründeki dağıtım, çiftçiden toptancıya, toptancıdan perakendeciye ve oradan da son tüketiciye olmak üzere çok katmanlı bir yapıya sahiptir. Şehirlerde nüfusun yoğun olması sebebiyle gıda ürünlerine talep hayli fazladır. Bu nedenle hem sistemin karmaşık bir yapıya sahip olması hem de talebin fazla olması, tüm zinciri göz önünde bulundurup etkili ve sürdürülebilir bir dağıtım ağı tasarlamayı oldukça zor bir hale getirmektedir. İnternet ağının hızla gelişmesi, internet üzerinden yapılan alışverişlerin artmasına ve e-ticaret kavramının ortaya çıkmasına neden olmuştur. İnternet üzerinde yapılan alışverişlerde ürünlerin teslimatı müşterinin bulunduğu yere ya da istediği bir yere yapıldığından bu durum müşteri açısından bir avantaj yaratmaktadır. Sipariş edilen bir ürünün müşterinin bulunduğu yere teslim edilmesi şeklinde tanımlanan Son Kilometre Teslimatı lojistik zinciri içerisinde yüksek maliyete sahip bir işlemdir. Ayrıca trafik sıkışıklığı, hava kirliliği, gürültü kirliliği gibi çevreye olumsuz etkileri de bulunmaktadır. Kentsel bölgelerde gıda ürünlerini son tüketiciye ulaştırırken trafikte kaybedilen zaman ve müşterilerin dağınık haldeki lokasyonları ürünlerin bozulmasına neden olabilmektedir. Bu nedenle ürünleri daha az maliyet ile daha hızlı taşınmasını sağlayacak dağıtım modelini kurmak daha zor hale gelmektedir.Bu çalışmada, raylı ulaştırma hattını dağıtım ağına entegre ederek ve kentlerde yoğunlaşmış olan küçük bakkal ya da market noktalarıyla işbirliği yaparak gıda ürünlerinin son tüketiciye ulaştırıldığı yeni bir kentsel dağıtım modeli önerilmektedir. Problem kısmi kapsamanın olduğu maksimum yerleşim problemi şeklinde tanımlanmıştır. Belirlenen yerleşim bölgesinde aday bakkal noktaları olduğu ve müşterilerin taleplerinin bu aday bakkallar tarafından karşılandığı varsayılmaktadır. Problemde tren istasyonları, yerleşim yerlerinde bulunan bakkallar ve son tüketiciler yani müşteriler olmak üzere üç hiyerarşik düzey bulunmaktadır. Raylı hattın başlangıç noktasına yakın bir noktada bulunan depodan bakkaların atandığı istasyona raylı hat ile bakkallara gıda ürünlerinin gönderildiği varsayılmaktadır. Problem iki amaca hizmet edecek şekilde formüle edilmiştir. Birinci amaç, tanımlanan bir kapsama parametresi ile müşterilerin en yakındaki bakkaldan taleplerini karşılamalarını sağlayarak talebin karşılanmasını maksimize etmektir. İkinci amaç ise tren istasyonları ile bakkallar arasındaki mesafeyi en aza indirmektir. İstanbul ili için sürdürülebilir bir optimizasyon yaklaşımı önerilmektedir. Çalışmada Marmaray hattı ve bu hatta entegre olarak çalışan M4 metro hattı kullanılmıştır. Ürünlerin metro ve Marmaray sistemlerinde taşınmasının, yolcu taşımacılığının başladığı en erken saatten daha önce yapıldığı varsayılmaktadır. Kurulan modeli analiz etmek için Türkiye'deki büyük bir bankanın müşterilerine ait kredi kartı işlem verileri kullanılmıştır. 0-1 tamsayılı programlama formülasyonu geliştirilerek veri setinden alınan küçük bir problem örneği GAMS programında çözülmüştür. Daha sonra gerçek veriler kullanılarak MATLAB 2019a programı ile problem çözdürülmüş ve optimum sonuç elde edilmiştir. Çalışmanın sonucunda amaç fonksiyonunu optimize edecek bakkal konumları belirlenmiştir. Belirlenen bu bakkalların konumları harita üzerinde gösterilmiştir.

In recent years, internet usage has become widespread with the rapid development of technology. In ten years, the amount of internet users has increased significantly. As a result of this increase, companies have become aware of the potential customer over internet. The fact that the Internet is a part of people's lives has forced some companies in many sectors to change their way of doing business. For this reason, companies have started to offer products or services to consumers over the internet in order to keep up with these new conditions and to maintain their presence in a developing competitive environment. It offered some companies the opportunity to use them as an alternative and effective way to reach their target markets. As a result, the concept of e-commerce has evolved and continues to grow rapidly. With the development of e-commerce, the process of delivering the products which are the last leg of the supply chain to the customers' place has become a big problem for the enterprises. This process, called the last mile delivery, is a very costly process with little efficiency. For this reason, firms have to make their supply chains effective to adapt to the new conditions that are formed in order to deliver the products to the end costumer more quickly with less cost. This makes it difficult to design a sustainable and effective distribution network within the supply chain.In addition to high cost and inefficiency, the last mile delivery has a negative impact on the environment. In cities, especially in metropolitan cities, both the number of people and the number of vehicles are quite high. This situation causes people to spend long hours in traffic, to create air pollution, to cause people to live in stress. In addition to this situation vehicles used to deliver products to the customer also cause negative effects such as traffic congestion in cities, air pollution resulting from increased CO2 emissions and noise pollution. This leads to a decrease in the quality of life in cities. In recent times, environmental awareness and responsibility have emerged in humans and both the state and the companies make investments that respect the environment. The concept of sustainability, which means ensuring the continuity of diversity and productivity of biological systems, has become a notable issue. Nowadays, electric vehicles are being used instead of fossil fuel powered vehicles. When the studies in the literature are analyzed, creative solutions have been presented to reduce the negative effects of the last mile delivery and the distribution of the product within the city as well as to protect the environment. Railway lines, electric vehicles, bicycle couriers are some of the examples presented to reduce the negative effects of last mile delivery.Especially in the food sector which is the subject of this study, is commonly multi-layered starting from the farmer to the wholesaler, then to logistics service provider, to the retailer and finally to the end-customer Thus, an effective and sustainable distribution network design considering the whole chain is quite complex and challenging. With the increasing use of internet and e-tailing, last-mile delivery has also become one of the challenges of grocery distribution. When the total number of deliveries made with the orders given over the internet according to the type of product is examined, the sales volume of the food products is quite high. It is also a critical issue that the distribution of food products is made faster than the danger of degradation of products. Therefore there exist many improvement areas in the traditional grocery distribution such as reducing the carbon emissions by better routing, changing the transportation modes or eliminating the food spoil during the distribution. The heavy traffic in the cities and the scattered location of the customers in the residential area cause a loss of time in the distribution of the products. Because of the danger of product deterioration, the time spent in traffic can increase the cost.In this study, we propose a new urban distribution model for grocery goods particularly by integrating the railway mode in the distribution network and enhancing the last mile delivery via collaboration with the small independent grocery shops dispersed densely around the urban area. The problem is defined as a maximal covering facility location problem with coverage parameter. The coverage parameter is explained as follows. Demand points that are within the minimum critical distance to a facility are said to be covered, points that are further than the maximum critical distance are said to be uncovered, and the points that are located between the minimum critical distance and the maximum critical distance are partially covered. The distribution network considered in this problem consists of three levels, that are (i) railway stations, (ii) small grocery stores and (iii) end-customers. It is assumed that there are candidate grocery points within a given residential area and that the demand of the customers can be divided. It is assumed that the railway is used to replenish the small grocery stores from a grocery warehouse which is close to the starting point of the railway line. The grocery goods are sent to from this warehouse to the specific railway stations to which grocery stores are assigned to and the grocery goods are picked up by the grocery stores. Contrary to the common practice of using trucks for the long-distance distribution process, using rail line as proposed in this study would contribute to cost-effectiveness and the carbon emission reduction even though, the latter is not explicitly added in the mathematical model. The network is also designed by incorporating the small grocery stores in order to maximize the coverage of the end-customers and to minimize the collecting cost of goods from the railway stations. The problem is formulized as a mixed integer programming model. The first objective is to maximize the fulfilment of the demand by ensuring that customers receive the service from the nearest grocery stores with the specified coverage parameter. The second is to minimize the distance between railway stations and grocery stores.A sustainable approach is proposed for a case study problem in Istanbul. The urban railway to be used is a metro line named as `Marmaray`. The M4 metro line which is integrated to the Marmaray line was also used. It is assumed that rail transportation systems are utilized for goods transportation before starting the daily scheduled earliest passenger transportation trip. To analyse the grocery demand in Istanbul a credit card transaction data in the category of grocery from a large Turkish bank is used. In this study, the locations of customers in Istanbul were used in the data set since Istanbul was selected as the application area. Candidate grocery points were created randomly in Istanbul. A total of 5159 customers, 200 candidate grocery stores and 22 stations were used in the study. The mathematical model was first solved by GAMS with the help of a small sample taken from the data set. After testing the accuracy of the model, the problem is solved by using the Optimization Toolbox of MATLAB 2019a program for a large-scale problem.As a result of the study, grocery points where the above-stated objectives for the distribution of food products are optimally provided have been identified. Information about which candidate grocery points will be used can be obtained from the solution of the problem. These determined grocery points are shown on the map. According to the results obtained, it is aimed to reduce the negative economic, social and environmental impacts that occur during the delivery of food products in the city.