在現在社會上,設施規劃問題(Facility Layout Problem, FLP)不斷的出現,小至工廠、公司的設施規劃,大至國家設施規劃。而設施規劃這議題也已經討論非常久,從一剛開始靜態設施規劃(Static Facility Layout Problem, SFLP)慢慢延伸至現在的動態設施規劃(Dynamic Facility Layout Problem, DFLP)。在過去的文獻大多只談到一致部門大小之靜態設施規劃或是不一致部門大小之靜態設施規劃甚至一致部門大小之動態設施規劃,在不一致部門大小之動態設施規劃方面的文獻可說是相當少。由於在現實的情況,不管是部門或是設施都屬於不一致面積大小,因此為了符合現實情況,本研究將對於不一致部門大小之動態設施規劃進行探討。 本研究主要利用彈性區帶架構(Flexible Bay Structure,FBS)進行不一致部門之設施規劃,在設置部門或設施時也會將走道也一併設置完成。在成本上的考量方面,不一致部門大小面積的動態設施規劃過去的文獻,考量部門之間搬運的方式都是以中心點對中心點搬運,但是在現實情況下有走道的設置,不可能直接無視於牆壁或隔間而直接進行搬運,因此本研究提出以出入口點的設置進行搬運,在重新安置成本(Rearrangement cost)的計算則是計算將上一期的部門沿著該期的出入口設置搬移至下一期預計的地點的移動距離。 由於設施規劃問題屬於NP-hard問題,因此本研究利用粒子群演算法(Particle Swarm Optimization,PSO)所求得解與登山法(Hill Climbing)進行比較,在實驗結果上顯示出在求解上皆比登山法來得佳。並且將本研究之方法應用於實際案例上,對於案例公司進行三期之設施規劃。在結果上顯示本研究之方法可以應用於現實情況上。
Nowadays, facility layout problems exist in places like companies or even countries. In the past, researches are conducted on Static Facility Layout Problem (SFLP); later, the focus has been shifted to Dynamic Facility Layout Problem (DFLP). In the past, most researches are about Equal-Area Static Facility Layout Problem (EA-SFLP) or Unequal-Area Static Facility Layout Problem (UA-SFLP) and Equal-Area Dynamic Facility Layout Problem (EA-DFLP). A few journal articles are published about Unequal-Area Dynamic Facility Layout Problem (UA-DFLP). In reality, generally the areas of facilities and departments are unequal in sizes. Therefore, this research will be conducted on Unequal-Area Dynamic Facility Layout Problem to better reflect the real situation. This research applies the method of Flexible Bay Structure (FBS) to solve the unequal area aspect of UN-DFLP with consideration of flow line for each department. Regarding the rearrangement cost, previous researches take the centroid-to-centroid approach for interdepartmental rearrangement. However, in reality there are aisles along each department; the consideration is part of the research. Because the problem is a NP-hard problem, we apply the Particle Swarm Optimization (PSO) to find the near-optimal solution and compare it with the Hill Climbing heuristic. Based on the result, PSO is proven to be better than Hill Climbing. A case study is also presented to illustrate the practical use of the proposed ideas and techniques.