حل مشکل برون سپاری ترکیبی محصول یکپارچه با استفاده از الگوریتم امپریالیستی رقابتی

The integrated product mix-outsourcing optimization is a major problem in manufacturing enterprise. Generally, heuristic or meta-heuristic solution approaches are used to optimize such problems. Heuristic approaches for these problems include Theory of Constraints (TOC) and Standard Accounting. Sometimes heuristic approaches are inefficient especially in large problems and instead, in these cases meta-heuristic algorithms have been applied extensively. In this paper a novel meta-heuristic algorithm “Imperialist Competitive Algorithm” (ICA) is applied to solve the integrated product mix-outsourcing optimization problem. Also, the results obtained from ICA are compared with the results of TOC and Standard Accounting approaches.

Maximizing the enterprise profit and fully utilizing the limited production resources is a key optimization problem in the strategic planning of enterprises. Furthermore, another major difficulty is how to optimize the enterprise’s production throughput within the determined product types (Wang, Sun, Si, & Yang, 2008). Aggregating the above two optimization concerns results in the product mix optimization problem which involves identification of the type and quantity of product to produce in order to make the most profit (Coman & Ronen, 2000). In the competitive market enterprises try to meet the market demand as much as possible. Although, this increases the sale revenue and profit, but in most cases, the enterprise resources produce the total demand are insufficient. Therefore, the enterprises tend to outsource part of the demand to external suppliers. Hence, the product mix and the outsourcing decisions have to be made simultaneously in many of the real world cases and can be referred to as the integrated product mix-outsourcing optimization problem. Küttner (2004) stated that to solve the product mix problem and decide on the production volume for each product, the demand’s forecast and the value of utilized resources should be considered. Lea and Fredendall (2002) demonstrated that the choice of product mix influences on the enterprise performance measures such as profit, work-in-progress (WIP), customer service, and manageability of the shop. To solve the product mix problem, there are two general approaches in the literature: heuristic and meta-heuristic algorithms.

In this paper, we formulated the integrated product mix-outsourcing problem in LP format. Then, we applied a meta-heuristic algorithm called Imperialist Competitive Algorithm (ICA) for the first time to solve such problems. Then we examined ICA against another two well-known approaches in the literature (TOC and Standard Accounting) and obtained better solution. ICA is capable of solving large-scale integrated product mix-outsourcing problem. Furthermore, contrary to TOC and Standard Accounting methods, ICA considers all constraints inside and outside the enterprise. Availability and cost of outside resources as well as the resources in the enterprise are considered simultaneously. To name a few future research possibilities new constraints and multi-objective cases can be mentioned. In general, ICA has a promising potential to be used as a new solution approach in a variety of problems.