An algorithm for optimising the servicing of products with constrained, multiple defects

The twin pillars of sustainable development are the conservation of natural resources and the management of waste. Waste is generated whenever a product is serviced or repaired, or when it is ultimately discarded at the end of its useful life. In order to manage such waste, the servicing options and costs must first be ascertained. This paper presents an algorithm for the generation of optimal disassembly and re-assembly sequences for the servicing of products with multiple defects, subject to constraints such as inaccessible components. The multiple service action (MSA) algorithm determines the minimum total servicing cost for a product network based on Floyd’s Algorithm, a shortest path algorithm. Well-established shortest path algorithms, which compute the shortest route between any pair of nodes in a network, are unable to handle multiple defects. The product network is first constructed, depicting the components and sub-assemblies as nodes, and embodying in directed arcs, the labour, materials and tooling costs associated with disassembly and re-assembly, as well as the cost to repair, reuse, recycle or dispose of the defective components. The MSA algorithm was tested on seven different product networks representing multiple defective components that can be serviced by different feasible routes. For each feasible service route, associated costs were computed. It was established that the algorithm was able to generate optimum disassembly and re-assembly routes for the servicing of products with multiple defects subject to constraints.