摘 要
營建工程師往往面臨如何規劃工地設施，例如工地的出入口、假設工程、各作業的工作區域配置，及材料的堆積等，而使得工地各作業順暢，進而提昇整體工地的生產力。以往的相關研究中，著重於靜態的設施配置規劃，也因此設施配置僅能適用於較為單純的事前規劃，無法依專案時間點的不同做一完整的考量。此外在設施間的關係上多使用主觀的判斷，也容易造成設施配置時的偏頗。再者對於處理不合理解的方法，多採用直接刪除或利用處罰函數，造成解答品質不穩定，及缺乏彈性供決策選擇。
本研究參考過去研究成果，提出改進辦法，將設施配置問題以專案的時程為基礎，依據設施與作業之關係及其產生的變化，建立一以時程為基礎的動態設施配置模式，並利用基因演算法來求解。
研究成果顯示，以專案時程為基礎的設施配置，不僅能較符合作業項目的需求，且利用權衡曲線為目標的求解模式能提供較具彈性的設施配置決策，此外本研究開發出一易於使用的電腦程式，以供工地工程師進行較便利的設施配置，作為本研究最後之具體成果。

Abstract
Construction engineers usually have to properly allocate construction facilities, such as material warehouses, the batch plant, or the job site office, so that the workflows within different working parties are smooth and the high productivity at the job site can be achieved. However, because of the dynamic sequences at the construction site, it is important that the facility layout of the construction site should be coordinated with the project schedule. Previous research related to the dynamic facility layout problem takes subjective weighting approach to determine the closeness between facilities. Such approach usually results in that the layout of the facilities is controlled by the subjective closeness coefficients, which is not practically acceptable. In addition, the solutions generated by the previous research are not flexible for the construction engineers to perform what-if analysis.
This research first analyzed the characteristics of different construction facilities during the construction period in terms of variations in size and location. A new dynamic facility layout model that the closeness between facilities is determined according to the project schedule is then developed. Furthermore, an elite non-dominate genetic algorithm (ENDGA) based on Pareto front is developed to solve the newly established model. Results showed that with the objectively determining the closeness of facilities, the facility layout is more practically acceptable at the construction site. The solutions generated by using ENDGA also provide the more flexible facility layout. A user-friendly program coded with Microsoft Visual Basic 6, is built to provide a practical tool for the construction engineers to easy the facility layout planning process.

1.Bozer, Y. A., and Miller, R. D., and Erlebacher, S. J. (1994). “An Improvement-Type Layout Alogrithm for Single and Multiple Floor Facilities.” Management Science, 40(7), 918-932.
2.Camp, D. J. V., and Carter, M. W., and Vannelli, A. (1991). “A Nonlinear Optimization Approach for Sloving Facility Layout Problems.” European Journal Of Operational Research, 57, 174-189.
3.Davidson, J. W. (1999). “Evolution Program For Geometry of Rectilinear Looped Networks.” J. Comp. in Civ. Engrg., ASCE, 13(4), 246-252.
4.Elbeltagi, E., and Hegazy, T., and Hosny, A. H., and Eldosouky, A. (2001). “Schedule-Depentend Evolution of Site Layout Planning.” Construction Management and Economics, ASCE, 19, 689-697.
5.Feng, C. W., and Liu, L., and Burns, S. A. (1997). “Using Genetic Algorithms to Solve Construction Time-Cost Trade-Off Problems.” J. Comp. in Civ. Engrg., 11(3), 184-189.
6.Foulds, L. R., and Robinson, D. F. (1978). ”Graph Theoretic Heuristics for the Plant Layout Problem.” International Journal of Production Research, 16(1), 27-37.
7.Hassan, M. M. D., and Hogg, G. L., and Smith, D. R. (1986). “SPIRAL: A Construction Alogrithm for Area Placement Evaluation.” International Journal of Production Research, 24, 1283-1295.
8.Hegazy, T., and Elbeltagi, E. (1999). “EVOSIT: Evolution-Based Model for Site Layout Plaaning.” J. Comp. in Civ. Engrg., ASCE, 13(3), 198-206.
9.Li, H., and Love, P. (2000). “Genetic Search for Solving Construction Site-level Unequal-area Facility Layout Problems.” Automation In Construction, 9, 217-226.
10.Li, H., and Love, P. (1998). “Site-level Facilities Layout Using Genetic Algorithms.” J. Comp.in Civ. Engrg., ASCE, 12(4), 227-231.
11.Li, H., and Love, P. (1997). “Using Improved Genetic Alogrithms to Facility Time-Cost Optimization.” J. Con. Engrg. Mgmt., ASCE, Sept 233-237.
12.Meller, R. D., and Gau, K. Y. (1996). “The Facility Layout Problem: Recent and Emerging Trends and Perspectives.” Journal of Manufacturing Systems, 93(5), 351-366.
13.Rosenblatt, M. J. (1986). “The Dynamics of Plant Layout.” The Institute of Management Sciences, 32(1), 76-85.
14.Tam, C. M., and Tong, K. L., and Chan, K.W. (2001).”Genetic Alogrithm for Optimizing Supply Locations Around Tower Crane.” J. Con. Engrg. Mgmt., ASCE, 127(4), 315-321.
15.Tam, K. Y. (1992). “A Simulated Annealing Alogrithm for Allocating Space to Manufacturing Cells.” International Journal of Production Research, 30, 63-87.
16.Tommelein, L. D., and Levitt, R. E., and Hayes-Roth, B., and Confrey, T. (1991). “SightPlan experiments: Alternative strategies for site layout design.” J. Comp. Civ. Engrg., ASCE, 5(1), 42-63.
17.Tsuchiya, K., and Bharitkar, S., and Takefuji, Y. (1996). “A Neural Netwook Approach to Facility Layout Problems.” European Journal of Operational Research, 89, 556-563.
18.Warszawski, A., and Peer, S. (1973). “Optimizing the location of facilities on a building site.” Operat. Res., Quart, 24(1), 35-44.
19.Yeh, I. C. (1995). “Construction site layout using annealed neutral network.” J. Comp. in Civ. Engrg., ASCE, 9(3), 201-208.
20.Zouein, P. P., and Tommelein, L. D. (2001). “Improvement Alogrithm For Limited Space Scheduling.” J. Con. Engrg. Mgmt., ASCE, 125(6), 116-124.
21.Zouein, P. P., and Tommelein, L. D. (1999). “Dynamic Layout Planning Using A Hybrid Incremental Solution Method.” J. Con. Engrg. Mgmt., ASCE, 125(6), 400-408.
中 文 部 份
22.呂守陞、楊崇揮，「遺傳演算法在資源限制下工期成本交易最適化排程模式運用之研究」，民國八十八年，第十一卷，第三期，第559-565頁
23.馬曉峰、馮重偉、鄭道明，「基因演算法在工程設施配置問題之應用」，八十九年「第四屆營建工程與管理研究成果聯合發表會論文集」， 2000，第120-130頁。
24.葉怡成，「以遺傳演算法作設施工址配置」，行政院國家科學委員會專題研究計劃成果報告，1995。
22.張宏旭，基因演算法在設施配置規劃上之應用，國立成功大學土木工程學系碩士論文，民國90年。
23.蘇木春、張孝德，機器學習：類神經網路、模糊系統以及基因演算法法則，全華科技圖書股份有限公司，第9.1-9.12頁。