Jump to Main Content
Mobile robots and evolutionary optimization algorithms for green supply chain management in a used-car resale company
- V. Sathiya, M. Chinnadurai, S. Ramabalan, Andrea Appolloni
- Environment, development and sustainability 2021 v.23 no.6 pp. 9110-9138
- algorithms, automation, carbon, environment, greenhouse gases, guidelines, models, supply chain, supply chain management, transportation, India
- To ensure environment friendly products in the international supply chain scenario, an important initiative is reverse supply chain (RSC). The benefits (environmental and financial) from a RSC are influenced by disposal of reusable parts, cost factors and emissions during transportation, collection, recovery facilities, recycling, disassembly and remanufacturing. During designing a network for reverse supply chain, some objectives related to social, economic and ecological concerns are to be considered. This paper suggests two strategies for reducing the costs and emissions in a network of RSC. This research work considers design of RSC for a used-car resale company. First strategy outlines the design of a mobile robot—solar-powered automated guided vehicle (AGV) for reducing logistic cost and greenhouse gas (GHG) emissions. The second strategy proposes a new multi-objective optimization model to reduce the costs and emissions of GHG. Strict carbon caps constraint is used as a guideline for reducing emissions. The proposed strategies are tested for a real-world problem at Maruti True Value network design in Tamil Nadu and Puducherry region of India. Two algorithms namely Elitist Nondominated Sorting Genetic Algorithm (NSGA-II) and Heterogeneous Multi-Objective Differential Evolution algorithm (HMODE) are proposed. HMODE is a new improved multi-objective optimization algorithm. To select the best optimal solution from the Pareto-optimal front, normalized weighted objective functions (NWOF) method is used. The strength or weakness of a Pareto-optimal front is evaluated by the metrics namely ratio of non-dominated individuals (RNI) and solution spread measure (SSM). Also, Algorithm Effort (AE) and Optimiser Overhead (OO) are utilized to find the computational effort of multi-objective optimization algorithms. Results proved that proposed strategies are worth enough to reduce the GHG emissions and costs.