A multi-objective closed-loop supply chain model of lead acid batteries considering the environmental impacts

• Main Authors: Lao, Mark Ian Y., Tan, Charles Andrew C., Valderrama, Michael Anthony J.
• Format: text
• Language: English
• Published: Animo Repository 2007
• Subjects: Lead-acid batteries; Storage batteries; System analysis; Environmental impact analysis; Production engineering
• Online Access: https://animorepository.dlsu.edu.ph/etd_bachelors/8139
• Institution: De La Salle University

The global market for lead acid batteries amounted to $17.45 million a few years ago and is still steadily growing. The problem with lead acid batteries comes in its recycling and manufacturing processes. These processes emit dangerous substances which significantly affect the climate, human health, marine life and other environmental factors. Studies in the past have focused mostly on the economic objectives of the lead acid battery supply chain. Some authors have proposed including the environmental impacts of the chain but non have incorporated an assessment tool that computers the environmental impacts alongside the computation of the economic objectives. A mixed integer nonlinear programming model was formulated for a closed loop supply chain model of lead acid batteries. The closed loop supply chain has three echelons for the forward flow and four echelons for the reverse flow. In the forward flow the echelons are the manufacturing plants, warehouses, and the retailers. In the backward flow, the echelons included are the retailers, warehouses, lead recycling facilities, and the manufacturing plants. The ISO developed impact assessment methodology was used to quantify the environmental impacts of all the processes in the chain. The model has two objectives, minimizing the economic cost and environmental impacts. The major decision variables include site selection, product allocation, and treatment installation. In the model validation, the mixed integer nonlinear programming model was linearized for ease of solving. The model was translated to GAMS language and solved with the CPLEX algorithm. The generating approach was used in validating the model. Solution closer to the optimal were generated in each run by adjusting the aspiration levels until the terminating rule was satisfied. In the sensitivity analysis, designed experiments were used to analyze relationships between the parameters and the following responses: the economic costs, the environmental objectives, and the amount of batteries flowing from the manufacturing plant and the lead recycling plants in both the urban and rural areas, and the amount of virgin lead purchased. It was found through the 2K factorial design that the demand, recycling cost, weight for human toxicity in the urban area and rural area, and the weight for marine toxicity in the rural area are the significant factors that affect the response. The Response Surface Methodology was used to analyze the effects of the significant factors on the decision variables and objective function. The location of facilities and allocation of products were affected most by the demand, recycling cost, weights of human toxicity in the urban area and marine toxicity in the rural area. Increasing the weight of human toxicity in the urban area would make it very favorable for the model to locate the lead recycling and manufacturing facilities in the rural area. On the other hand, increasing the weight of marine toxicity in the rural area would make to model locate the emission causing facilities in the urban area. Increasing the recycling costs would make recycling used batteries less favorable, and cause the model to use a higher percentage of virgin lead in order to satisfy the demand. Recommendations for further study include modeling the pricing strategies that would increase the rate of returns. Second, the use of Environmental Protection System, a possible alternative to the impact assessment used in this study, may be explored.