Three-echelon Supply Chain Network Research Articles

A local supply chain inventory planning with varying perishability rate product: A case study

Motivated by a real-life case in the dairy supply chain, we study the management of deteriorating inventories in an existing three-echelon supply chain network with a local retailer with multiple customers. Like many industries, the focal company categorizes the customers into two different clusters in the supply chain downstream with different, uncertain demand functions: (i) many, small, ordinary customers; (ii) few, large, premier customers. The focal company intends to determine the optimal inventory policy while the deterioration rate of the product is time sensitive. The problem is formulated using two concepts: 1) queuing theory and finite-horizon Semi-Markov process to characterize the structure of the optimal inventory policy; 2) integrated inventory system in the supply chain at the network level. We formulate the problem mathematically and show that the model is nonlinear and nonconvex. Therefore, the model cannot be solved by using commercial optimization software products to reach global optimal. Consequently, two solution approaches are designed to solve the problem: (i) a modified version of the Invasive Weed Optimization algorithm, named adaptive Invasive Weed Optimization, which is designed, tuned and validated (using well-known algorithms) to tackle the problem; (ii) an Adaptive Heuristic Method. We show that adaptive Invasive Weed Optimization is superior in the quality of solutions while Adaptive Heuristic Method is faster. Various sensitivity analyses applied to the real data show when facing both perishability and uncertainty, the effect of perishability outweighs the effect of uncertainty.

Complexity in three-echelon supply chain network and manufacturing firm’s operational performance

The purpose of this paper is to understand the combined impact of supply chain structural complexity and operational complexity on the operational performance of manufacturing firms in a three-echelon supply chain network. The paper uses Shannon’s information entropy theory and agent-based modeling to find the intended relationships. The structural complexity is represented by the numbers of customers and suppliers associated with the focal manufacturing firm, whereas the operational complexity is represented by demand variability experienced by the manufacturing firm. The operational performance of the manufacturing firm is measured in terms of its capacity utilization, demand fulfillment, and inventory level. The paper finds that increasing the number of customers reduces capacity utilization and demand fulfillment, and increases inventory level whereas increasing the number of suppliers increases capacity utilization, demand fulfillment, and inventory level for the focal manufacturing firm. The results also show that increasing the number of customers and suppliers together initially improves the capacity utilization and demand fulfillment of the manufacturing firm but these deteriorate subsequently. Furthermore, increasing operational complexity reduces the operational performance of the firm. Thus, the paper finds how the operational performance of the focal manufacturing firm changes with the changes in supply chain structural complexity and operational complexity in a three-echelon supply chain. The results imply that manufacturing firms need to carefully orchestrate their supply chain network as increasing the number of customers and suppliers in the network does not necessarily improve their operational performance.

Designing a locating-routing three-echelon supply chain network under uncertainty

PurposeIn the present research, location and routing problems, as well as the supply chain, which includes manufacturers, distributor candidate sites and retailers, are explored. The goal of addressing the issue is to reduce delivery times and system costs for retailers so that routing and distributor location may be determined.Design/methodology/approachBy adding certain unique criteria and limits, the issue becomes more realistic. Customers expect simultaneous deliveries and pickups, and retail service start times have soft and hard time windows. Transportation expenses, noncompliance with the soft time window, distributor construction, vehicle purchase or leasing, and manufacturing costs are all part of the system costs. The problem's conceptual model is developed and modeled first, and then General Algebraic Modeling System software (GAMS) and Multiple Objective Particle Swarm Optimization (MOPSO) and non-dominated sorting genetic algorithm II (NSGAII) algorithms are used to solve it in small dimensions.FindingsAccording to the mathematical model's solution, the average error of the two suggested methods, in contrast to the exact answer, is less than 0.7%. In addition, the performance of algorithms in terms of deviation from the GAMS exact solution is pretty satisfactory, with a divergence of 0.4% for the biggest problem (N = 100). As a result, NSGAII is shown to be superior to MOSPSO.Research limitations/implicationsSince this paper deals with two bi-objective models, the priorities of decision-makers in selecting the best solution were not taken into account, and each of the objective functions was given an equal weight based on the weighting procedures. The model has not been compared or studied in both robust and deterministic modes. This is because, with the exception of the variable that indicates traffic mode uncertainty, all variables are deterministic, and the uncertainty character of demand in each level of the supply chain is ignored.Practical implicationsThe suggested model's conclusions are useful for any group of decision-makers concerned with optimizing production patterns at any level. The employment of a diverse fleet of delivery vehicles, as well as the use of stochastic optimization techniques to define the time windows, demonstrates how successful distribution networks are in lowering operational costs.Originality/valueAccording to a multi-objective model in a three-echelon supply chain, this research fills in the gaps in the link between routing and location choices in a realistic manner, taking into account the actual restrictions of a distribution network. The model may reduce the uncertainty in vehicle performance while choosing a refueling strategy or dealing with diverse traffic scenarios, bringing it closer to certainty. In addition, two modified MOPSO and NSGA-II algorithms are presented for solving the model, with the results compared to the exact GAMS approach for medium- and small-sized problems.

Bi-objective green supply chain network design under disruption risk through an extended NSGA-II algorithm

Green supply chain network design plays an important role in reducing environmental pollution and resource consumption. Furthermore, emergency events may cause disruptions and seriously reduce the performance of entire supply chain. This paper considers a three-echelon supply chain network consisting of multi-manufacturers, retailers with disruption risks and capacity constraints, and consumers with green preference. Different from traditional supply chain network design, this paper considers respectively two objectives from the manufacturers and consumers with green preference in selecting appropriate retailers under disruption risks. Moreover, we propose an extended elitist non-dominated sorting genetic algorithm (NSGA-II) to solve the bi-objective optimization model. The extended algorithm is considered to combine original NSGA-II algorithm with the migration operation of the biogeography-based optimization (BBO) algorithm, thus leading to enrich the diversity of the population and improve the ability of global searching. Compared with the traditional NSGA-II algorithm, the extended NSGA-II algorithm can obtain more Pareto solutions with higher efficiency. Furthermore, the result shows a trade-off relationship between the manufacturer cost and consumer cost. Therefore, it is necessary to separate the two objectives of manufacturers and consumers in green supply chain network design.

Time-To-Recovery Prediction in a Disrupted Three-Echelon Supply Chain Using LSTM*

This paper introduces a Long-Short Term Memory (LSTM) neural network-based method for Time-To-Recovery (TTR) prediction within a cognitive digital supply chain twin framework to enhance supply chain resilience and improve decision-making under disruption. The introduced method is applied to a virtual three-echelon supply chain network modelled using discrete event simulation. The virtual system was used to train an LSTM neural network model to predict TTR under several disruption scenarios. Results show that predicted TTR values tend to be relatively lower than the actual values at early disruption stages, then improve throughout the progression of the disruption effect on the supply chain network.

A reliable location-inventory-routing three-echelon supply chain network under disruption risks

PurposeThis study aims to design a reliable multi-level, multi-product and multi-period location-inventory-routing three-echelon supply chain network, which considers disruption risks and uncertainty in the inventory system.Design/methodology/approachA robust optimization approach is used to deal with the effects of uncertainty, and a mixed-integer nonlinear programming multi-objective model is proposed. The first objective function seeks to minimize inventory costs, such as ordering costs, holding costs and carrying costs. It also helps to choose one of the two modes of bearing the expenses of shortage or using the excess capacity to produce at the expense of each. The second objective function seeks to minimize the risk of disruption in distribution centers and suppliers, thereby increasing supply chain reliability. As the proposed model is an non-deterministic polynomial-time-hard model, the Lagrangian relaxation algorithm is used to solve it.FindingsThe proposed model is applied to a real supply chain in the aftermarket automotive service industry. The results of the model and the current status of the company under study are compared, and suggestions are made to improve the supply chain performance. Using the proposed model, companies are expected to manage the risk of supply chain disruptions and pay the lowest possible costs in the event of a shortage. They can also use reverse logistics to minimize environmental damage and use recycled goods.Originality/valueIn this paper, the problem definition is based on a real case; it is about the deficiencies in the after-sale services in the automobile industry. It considers the disruption risk at the first level of the supply chain, selects the supplier considering the parameters of price and disruption risk and examines surplus capacity over distributors’ nominal capacity.

Solving a Multi-Item Supply Chain Network Problem by Three Meta-heuristic Algorithms

The supply chain network design not only assists organizations production process (e.g.,plan, control and execute a product’s flow) but also ensure what is the growing need for companies in a longterm. This paper develops a three-echelon supply chain network problem including multiple plants, multiple distributors, and multiple retailers with amulti-mode demand satisfaction policy inside of production planning and maintenance. The problem is formulated as a mixed-integer linear programming model. Because of its NP-hardness, three meta-heuristic algorithms(i.e., tabu search, harmony search and genetic algorithm) are used to solve the given problem. Also, theTaguchi method is used to choose the best levels of the parameters of the proposedmeta-heuristic algorithms. The results show that HS has abetter solution quality than two other algorithms.

Multi-objective optimization for a combined location-routing-inventory system considering carbon-capped differences

<p style='text-indent:20px;'>A combined location-routing-inventory system (CLRIS) in a three-echelon supply chain network is studied with environmental considerations. Specifically, a bi-objective mixed integer programming model is formulated for the CLRIS to deal with the trade-offs between the total cost and the carbon-capped difference (CCD). A multi-objective particle swarm optimization (MOPSO) heuristic solution procedure is developed and implemented to solve the bi-objective mixed integer programming problem. The bi-objective mixed integer programming model and the MOPSO heuristic procedure are applied to a real-life problem as an illustrative example. The approximate nondominated frontier formed by solutions not dominated by others can be used for the decision makers to make trade-offs between the total cost and the CCD. Sensitivity analyses are conducted, and the relationship among the carbon cap, CCD, the total cost and the carbon prices are examined, and relevant managerial insights are provided. Comparisons with other existing algorithms show that the MSPSO heuristic procedure has very good performance.</p>

Robust master planning of a socially responsible supply chain under fuzzy-stochastic uncertainty (A case study of clothing industry)

This paper addresses an integrated physical supply channel, aggregate production, and transportation planning problem through the lens of social sustainability under uncertainty. The three-echelon supply chain network includes multiple suppliers, a manufacturer and multiple distribution centres. Social-related criteria such as workers’ job-security and health (i.e. working conditions) and social investment are contemplated in the supply chain under consideration. In doing so, a chance constraint mixed integer non-linear programming model is built to determine the centralized planning of procurement-production-transportation under fuzzy-stochastic uncertainty over the tactical planning horizon. Moreover, an efficient hybrid solution procedure is developed utilizing convexification and defuzzification strategies. Finally, a real-case study in clothing industry is presented to show the model applicability and the solution procedure efficiency.

Optimization of Multi-period Three-echelon Citrus Supply Chain Problem

In this paper, a new multi-objective integer non-linear programming model is developed for designing citrus three-echelon supply chain network. Short harvest period, product specifications, high perished rate, and special storing and distributing conditions make the modeling of citrus supply chain more complicated than other ones. The proposed model aims to minimize network costs including waste cost, transportation cost, and inventory holding cost, and to maximize network’s profits. To solve the model, firstly the model is converted to a linear programming model. Then three multi-objective meta-heuristic algorithms are used including MOPSO, MOICA, and NSGA-II for finding efficient solutions. The strengths and weaknesses of MOPSO, MOICA, and NSGA-II for solving the proposed model are discussed. The results of the algorithms have been compared by several criteria consisting of number of Pareto solution, maximum spread, mean ideal distance, and diversification metric.Computational results show that MOPSO algorithm finds competitive solutions in compare with NSGA-II and MOICA.

Supply chain network design under the risk of uncertain disruptions

Facility disruptions in the supply chain often lead to catastrophic consequences, although they occur rarely. The low frequency and non-repeatability of disruptive events also make it impossible to estimate the disruption probability accurately. Therefore, we construct an uncertain programming model to design the three-echelon supply chain network with the disruption risk, in which disruptions are considered as uncertain events. Under the constraint of satisfying customer demands, the model optimises the selection of retailers with uncertain disruptions and the assignment of customers and retailers, in order to minimise the expected total cost of network design. In addition, we simplify the proposed model by analysing its properties and further linearise the simplified model. A Lagrangian relaxation algorithm for the linearised model and a genetic algorithm for the simplified model are developed to solve medium-scale problems and large-scale problems, respectively. Finally, we illustrate the effectiveness of proposed models and algorithms through several numerical examples.

Warehouse redesigning in a three-echelon supply chain network with consideration of routing under uncertainty

This paper addresses the problem of redesigning a three-echelon supply chain network under uncertainty. Since one of the most realistic problems that supply chains are dealt with is routing of vehicles, routing constraints with a split delivery condition are considered in our proposed model. Also, the possibility of outsourcing is considered in order to satisfy demands that exceed the production capacity. Furthermore, in order to deal with the presence of uncertainty in the problem, a light robust approach is developed. The performance of the proposed model is illustrated using a simulation procedure. Sensitivity analysis on the proposed model is also presented in the paper. The results show that the proposed method has a better performance than Light Robust approach and can be used as a useful managerial tool in redesign problems.

Complex network oriented artificial bee colony algorithm for global bi-objective optimization in three-echelon supply chain

Finding the best flow patterns (i.e., choices of resources) for a family of products is a key part of supply chain management. It primarily focuses on reasonable selecting suppliers for every component, selecting plants for assembling every sub- or final assembly, and selecting the delivery options to bring products to customers. Different selecting operations form different cost and lead-time. Balancing a trade-off between cost and lead-time is a non-trivial problem in a three-echelon supply chain, which forms a complex network. We focus on finding the best flow patterns in which reasonable selections can be formed together to provide products or services. The objective is to minimize the bi-objective of cost and lead-time for any product. In this paper, we propose a complex network oriented artificial bee colony algorithm, which can be processed in parallel, to tackle the so-called combinatorial problem. Besides, we employ simulated annealing and gradient descent to find global Pareto optimal solutions in a supply chain network. Extensive experiments on the three-echelon supply chain network demonstrate the superiority of our proposals: (1) the proposed CN-ABC and CN-ABC-SAGD have the capability of discovering global POS in a complex three-echelon SCN; (2) the speed of searching global POS is accelerated to satisfy the requirement of its complexity of a SCN.

Minimization of Logistics Cost and Carbon Emissions Based on Quantum Particle Swarm Optimization

This paper aims to simultaneously minimize logistics costs and carbon emissions. For this purpose, a mathematical model for a three-echelon supply chain network is created considering the relevant constraints such as capacity, production cost, transport cost, carbon emissions, and time window, which will be solved by the proposed quantum-particle swarm optimization algorithm. The three-echelon supply chain, consisting of suppliers, distribution centers, and retailers, is established based on the number and location of suppliers, the transport method from suppliers to distribution centers, and the quantity of products to be transported from suppliers to distribution centers and from these centers to retailers. Then, a quantum-particle swarm optimization is described as its performance is validated with different benchmark functions. The scenario analysis validates the model and evaluates its performance to balance the economic benefit and environmental effect.

Developing benders decomposition algorithm for a green supply chain network of mine industry: Case of Iranian mine industry

This paper attempts to design a three-echelon supply chain network for mine industry considering suppliers, manufacturing centers and distributors with different kinds of vehicles. A multi-objective mixed integer programming model is proposed to minimize transportation costs, shortage costs, purchase costs, production costs and CO2 emissions of vehicles with fuzzy demand values. ε-constraint method is used in order to convert multiple objective functions to a single objective function. Benders decomposition algorithm is applied to solve the model under three demand-based scenarios. A case study is done on Iranian mine industry to present the significance and applicability of the proposed model in a real-life case problem as well as the efficiency of Benders decomposition.

Tri-level decision-making for decentralized vendor-managed inventory

Vendor-managed inventory (VMI) is a common inventory management policy which allows the vendor to manage the buyer's inventory based on the information shared in the course of supply chain management. One challenge in VMI is that both the vendor and buyer are manufacturers who try to achieve an inventory as small as possible or even a zero inventory; it is therefore difficult to manage inventory coordination between them. This paper considers a decentralized VMI problem in a three-echelon supply chain network in which multiple distributors (third-party logistics companies) are selected to balance the inventory between a vendor (manufacturer) and multiple buyers (manufacturers). To handle this issue, this paper first proposes a tri-level decision model to describe the decentralized VMI problem, which allows us to examine how decision members coordinate with each other in respect of decentralized VMI decision-making in a predetermined sequence. We then turn our attention to the geometry of the solution space and present a vertex enumeration algorithm to solve the resulting tri-level decision model. Lastly, a computational study is developed to illustrate how the proposed tri-level decision model and solution approach can handle the decentralized VMI problem. The results indicate that the proposed tri-level decision-making techniques provide a practical way to design a novel manufacturer-manufacturer (vendor-buyer) VMI system where third-party logistics are involved.

Risk-pooling strategy, lead time, delivery reliability and inventory control decisions in a stochastic multi-objective supply chain network design

A three-echelon supply chain network design is considered and a multi-objective mixed integer non-linear problem is modeled. Risk-pooling, lead time between facilities and the concept of the reliability are used to modeling. The objectives of the model are minimization of the supply chain’s total cost and maximization the delivery reliability, which can be equal to minimize the risk of not delivering the products with regard to stochastic demand. To solve the model a heuristic hierarchical algorithm is presented. Numerical examples are considered to test and verify the Pareto front. The results show the effectiveness and efficiency of the heuristic algorithm.

Supply Chain Network Design under Demand Uncertainty and Supply Disruptions: A Distributionally Robust Optimization Approach

We develop a robust optimization model for designing a three-echelon supply chain network that consists of manufacturers, distribution centers, and retailers under both demand uncertainty and supply disruptions. The market demands are assumed to be random variables with known distribution and the supply disruptions caused by some of the facilities faults or connection links interruptions are formulated by several scenarios with unknown occurrence probabilities. In particular, we assume the probabilities that the disruption scenarios happen belong to the two predefined uncertainty sets, named box and ellipsoid uncertainty sets, respectively. Through mathematical deductions, the proposed robust SCN design models can be transformed into the tractable linear program for box uncertainty and into second-order cone program for ellipsoid uncertainty. We further offer propositions with proof to show the equivalence of the transformed problems with the original ones. The applications of the proposed models together with solution approaches are investigated in a real case to design a tea supply chain network and validate their effectiveness. Numerical results obtained from model implementation and sensitivity analysis arrive at important practical insights.

Optimizing a bi-objective multi-product multi-period three echelon supply chain network with warehouse reliability

Bi-objective optimization of a multi-product multi-period three-echelon supply chain network consisting of manufacturing plants, distribution centers (DCs) each with uncertain services, and customer nodes is aimed in this paper. The two objectives are minimization of the total cost while maximizing the average number of products dispatched to customers. The decision variables are: (1) the number and the locations of reliable DCs in the network, (2) the optimum number of items produced by plants, (3) the optimum quantity of transported products, (4) the optimum inventory of products at DCs and plants, and (5) the optimum shortage quantity of the customer nodes. The problem is first formulated into the framework of a constrained bi-objective mixed integer linear programming model. Then, to solve the problem using the GAMS software, six multi-objective decision-making (MODM) methods are investigated in order to select the best in terms of total supply chain cost, total expected number of products dispatched to customers, and their required CPU time, simultaneously. At the end, some numerical illustrations are provided to show the applicability of the proposed methodology.

COALITION OR DECENTRALIZATION: A GAME-THEORETIC ANALYSIS OF A THREE-ECHELON SUPPLY CHAIN NETWORK

Supply chains have become the major and dominant paradigm of business and competition. The main challenge is how to act in multi-echelon supply chains considering the levels involved. Making a choice independently or integrating with some or all levels will be a critical decision, and therefore affects the overall profit of the chain. This article proposes a non-cooperative game theory approach to helping in making a better decision in the supply chain and gaining the most accessible benefit. Our research considers unlimited three-echelon supply chains with S suppliers, M manufacturers and K retailers. The Nash equilibrium and definition are used bearing in mind inventory and pricing and marketing cost as decision variables for this matter. This paper studies a three-echelon supply chain network and focuses on the value of integrating a pair of partners in the chain. In the decentralized case, the supplier sets its own price, the manufacturer points out order quantity, wholesale price and backorder quantity, and the retailer charges the final retail price of the product and marketing product. Though there are multiple players at a single echelon level, each manufacturer supplies only a specific product to a given retailer. In addition to the decentralized case, two integration scenarios have been taken into account: manufacturer-retailer and supplier-manufacturer. As for manufacturer-retailer integration, inventory/holding cost issues diminish to a single warehouse and the retailer does not have to enforce marketing effort any more. Supplier-manufacturer integration brings similar benefits. Under each scenario, all parties involved simultaneously set their strategies. Through a numerical experiment, 17 design cases (through designing experiments) have been developed and the total profit of the supply chain under each scenario has been evaluated. Statistical tests on the above introduced 17 experiments have found that the decentralized system performs significantly worse than the integration of the supplier with the manufacturer, whereas no significant difference can be observed regarding other combinations.