應用田口基因演算法於選擇和排程平衡專案組合

Hybrid Taguchi-genetic Algorithm for Selecting and Scheduling a Balanced Project Portfolio

對組織而言，專案組合選擇是一重要管理活動，也是個複雜決策過程。而兼顧專案組合平衡也是重要議題。因此，本研究所欲探討專案選擇和排程問題為選擇能滿足一組總淨現值最大的專案組合，而且必須兼顧在組織各策略領域的平衡以及在規劃期間內各專案何時開始以不違背年度預算限制額度。本論文所探討問題為兼顧在專案選擇、排程、平衡等三個重要議題，此問題在專案選擇實務中常見但少見於過去文獻。本論文我們建構此問題的0-1 整數規劃模式，並提出基因演算法（genetic algorithm, GA）求解。再者，結合田口方法以決定最佳參數水準以提昇所提GA 的效率。再透過一時小案例就GA 解與AMPL 所求得最佳解作比較以證實GA 的有效性，最後模擬許多大型評估所提GA 之績效。依運算結果得知，所提GA 演算法能有效率解決所提之問題。 The importance of maintaining a balancing portfolio makes the selection of projects a crucial management issue in many organizations and a complex decision-making process. In this study, we consider the problems associated with selecting and scheduling a set of R&D projects to maximize the overall net present value. The level of effort focused on each of the key strategies in the selected projects must be maintained, without violating the available annually budget. This study faces the problems frequently arising in real world applications, combining three issues (i.e., selection, scheduling, and balance) that have not been simultaneously addressed in previously suggested models. This paper proposes a zero-one integer programming model in conjunction with a genetic algorithm (GA) to overcome these problems. Furthermore, Taguchi Method was employed in the design of the GA parameters to increase the efficiency of the proposed method. A number of small problems were randomly generated to validate the effectiveness of the GA by comparing the solutions with those provided by AMPL. In addition, to evaluate the efficiency of the proposed GA six large problems were generated to identify the most appropriate parameters. From the computational results, we conclude that the proposed GA is capable of efficiently solving problems associated with the management of portfolios.