Multi-objective evolutionary approach for supply chain network design problem within online customer consideration

Abstract

Supply chain network design is one of the most important strategic decisions that need to be optimized for long-term efficiency. Critical decisions include facility location, inventory, and transportation issues. This study proposes that with a dual-channel supply chain network design model, the traditional location-inventory problem should be extended to consider the vast amount of online customers at the strategic level, since the problem usually involves multiple and conflicting objectives. Therefore, a multi-objective dual-channel supply chain network model involving three conflicting objectives is initially proposed to allow a comprehensive trade-off evaluation. In addition to the typical costs associated with facility operation and transportation, we explicitly consider the pivotal online customer service rate between the distribution centers (DCs) and their assigned customers. This study proposes a heuristic solution scheme to resolve this multi-objective programming problem, by integrating genetic algorithms, a clustering analysis, a Non-dominated Sorting Genetic Algorithm II (NSGA-II), and a Technique for Order Preference by Similarity to Ideal Solution (TOPSIS). Several experiments are simulated to demonstrate the possibility and efficacy of the proposed approach. A scenario analysis is conducted to understand the model's performance.

A scenario-based optimization model for planning and redesigning the sale and after-sales services closed-loop supply chain

“…Liao et al [18] developed a multi-objective dual-channel supply chain network model to extend the traditional location-inventory problem to consider online customers. They proposed a heuristic solution integrating the Genetic Algorithm, clustering analysis, and a Non-dominated Sorting Genetic Algorithm II to solve the problem.…”

Designing a dual-channel supply chain network considering dependent demand and discount

“…They applied profit index to evaluate the system's performance in different states and report their observations. Liao et al [19] formulated a multi-objective model for a dual-channel supply chain network. They aimed to; (1) minimize the facility location and inventory-related costs in the first objective function, (2) minimize the transportation cost in another objective function and (3) maximize the online customer's service to satisfy them in the final objective function.…”

Leagile supply chain network design through a dynamic two-phase optimization in view of order penetration point

“…There are different studies considering various features of supply chain in recent years [12,21,35,39]. Delivery time, quality and waste removal are the leanness's main criterion and delivery time and service level is the main criteria of agility.…”

“…In the proposed planning system, the model in phase I has to minimize variation effects of non-deterministic parameters, but since this phase covers strategic level of planning, it needs robust optimization. According to the theory of robust optimization stated in Ben-Tal et al [5] book, model should be sensitive to deviations from the optimal value in the OF (optimality robustness) and deviations from constraints feasibility (feasibility robustness) and for this, robustness cost should be mentioned in the OF of the model [21]. In model (5.6), cost of deviation from expected optimality is shown along with coefficient δ and penalty of deviations from feasibility is shown along with coefficients σ and λ.…”