A multi-objective decision making approach for dealing with uncertainty in EOL product recovery

Abstract

Recently, product recovery is considered as a promising approach to increase the economic benefit and reduce the environmental impact of products at the end-of-life (EOL) phase. After disassembly, each component of an EOL product can have different EOL options such as reuse, remanufacturing, recycling and landfill. Depending on the selected EOL options of components, the recovery value of the EOL product will be different. Thus, it is necessary to develop a decision-making method for EOL product recovery in order to select the best EOL options of components for maximizing the recovery value. However, EOL product recovery is usually characterized by a high level of uncertainty due to disassembly operations, product quality conditions and external attributes such as consumer preference, market requirement and price. Limited research has examined uncertainty of EOL product recovery during the EOL phase. Moreover, there has been a lack of research on dealing with uncertainty of EOL product recovery in a quantitative manner.To deal with this limitation, taxonomy of uncertainty metrics is developed through the whole EOL product recovery. The quantifications of uncertainty measures of EOL product recovery are formulated by different dimensions of quality condition, disassembly complexity and EOL recovery. A multi-objective decision making approach for dealing with uncertainty in EOL product recovery is proposed. Artificial bee colony (ABC) algorithm is employed to find the best EOL options of components with maximum feasibility and profit for EOL product recovery. A typical motor is used as a case study to illustrate the methodology. This paper addresses the uncertainty involved in determining the EOL options of components for EOL product recovery. The proposed approach closes a gap in the current EOL product recovery assessment criteria. By comparing to those selections of EOL options without considering uncertainty, the results show that considering uncertainty turns EOL product recovery more realistic and can give several good alternatives to decision makers.