船舶中管路佈置最為複雜的部分當屬機艙空間內各種設備間的聯繫，如何在有限空間內做到自動、省時、省材，以及符合船規的管路佈置，一直是造船技術研究中主要的發展方向，但隨著人因考量在現代工業中的角色日趨加重，船舶上人員的動態空間也逐漸被重視。
本研究以船舶機艙上設備操作度（Operability）考量，為佈管設計時的量化性限制條件（quantifiable constraints），搭配各種佈管上的規範，應用基因演算法軟體（GeneHunter）與電腦輔助設計軟體（PDMS）發展出一組兼具理論與實用性的佈管設計系統，在經濟成本與人因成本之間取得平衡，演算出最佳化的管路路徑。
本篇論文的基本架構為：
透過目前大型工程上通用的3D電腦輔助設計軟體(PDMS)為主要管路設計介面，發展出”設備在空間上虛擬體積界定”的模組，此處的用意為管路佈置時，除了必須閃避此空間中設備的真實體積，還必須考量設備操作空間的保留性，模組將產生佈管環境中各種設備的真實與虛擬體積的資料，提供路徑演算時約束性限制（restrictive constraints）與量化性限制（quantifiable constraints）條件的規範。
由於佈管最佳化設計屬於廣義型最佳化法則，無法求出目標函數的特性，因此本研究選用可以滿足上述兩種限制條件的演算法-基因演算法，此演算法的優點為同時間進行空間上多點的搜尋，並藉由懲罰函數的設定，可以於設計上顧慮到管路佈置中的許多細節，增加求得真正最佳解的可能性。

The most complicated part of pipeline arrangement in ship is the connection between equipments in engine room. It has been the main development in the research of shipbuilding technique that how to achieve automatically, time-saving, and material saving but according to ship’s rules at the finite space in the field of current shipbuilding industry. But the personnel's dynamic space design in ship’s engine room is taken gradually when the Human Factors became important in the modern industry.
This research is considering the operability of equipment in ship’s engine room as quantifiable constraint on the design of pipeline arrangement, and takes care of each kind of shipbuilding rules, to develop a pipeline design system with theory and practicability by the software applications of GeneHunter and PDMS. The system can achieve balance between cost and Human Factors, and calculates the optimization of pipeline arrangement.The basic structure of this thesis is:
It used the 3D charting software (PDMS) as the main interface, and developing a module which can judge the virtual space for equipment operation. The intention of this module is putting the consideration of equipment operability in a piping route design, and the module can create equipment’s data for restrictive constraints and quantifiable constraints in pipeline design.
Because the pipeline optimization design belongs to the generalized optimization principle, it is unable to determine the characteristic of fitness function. Therefore this research selected genetic algorithm that can satisfy above-mentioned constraints, and the advantage of this algorithm is that can simultaneous search different points and concern about many details for pipeline arrangement by penalty functions.

[1] Gen, M.; Cheng, R. Genetic Algorithm & Engineering Design, Wiley, 1997.
[2] Guiradello, R. “Optimization of process plant layout”. PhD Dissertation, University of Wisconsin: Madison, 1993.
[3] H Kimura and S Ikehira. “Automatic design for pipe arrangement considering valve operationality”. International Conference on Computer Applications in Shipbuilding. Shanghai, China Vol.2, pp. 121-126, 2009.
[4] Ito, T. “A genetic algorithm approach to piping route path planning”. Journal of Intelligent Manufacturing, 10, pp. 103–114, 1999.
[5] J. David Schaffer. “Multiple objective optimization with vector evaluated genetic algorithms”. Genetic Algorithms and their Applications: Proceedings of the First International Conference on Genetic Algorithms, pp. 93–100, 1985.
[6] Kamepalli, H.B. “The optimal basics for Gas”. IEEE Potentials, Vol.20, 2001.
[7] Johnson, J.M. and R.S. Yahya. “Genetic algorithms in engineering electromagnetic”. IEEE Antennas and Propagation Magazine, Vol.39, No.4, pp. 7-21, 1997.
[8] Lee, C. Y. “An algorithm for path connections and its applications”. IRE Transactions on Electronic Computers, EC-10, pp. 346-365, 1961.
[9] Murata, T.I., H. “MOGA multi-objective genetic algorithms, Evolutionary Computation”. IEEE International Conference on 1, pp. 289 , 1995.
[10]P. Hajela and C.Y. Lin. “Genetic search strategies in multicriterion optimal design”. Structural Optimization, pp. 99-107, 1992.
[11] Park J. and Storch R. “Pipe-routing Algorithm Development: Case Study of a Ship Engine Room Design”. Expert Syst. Appl, Vol. 23, pp. 299-309, 2002.
[12] Peter O’ Gray. “The Age of Modularity”. Adams and Steele publishers, 1999.
[13] Plant Design Management System（PDMS）User Manual. Version 11.6, England, 2004.
[14] Wangdahl, G. E., Pollock, S. M., & Woodward, J. B. “Minimum trajectory pipe routing”. Journal of Ship Research, 18(1), pp. 44-49, 1974.
[15] 王立嫻， “基因演算法應用於三維管路佈置之研究”，碩士論文，國立成功大學，造船暨船舶機械工程研究所，2001。
[16] 王順德，2002，“應用基因演算法於多層甲板結構物之最佳化設計”， 碩士論文，國立成功大學，造船暨船舶機械工程研究所。
[17] 行政院勞工委員會/勞工安全衛生研究所/勞工委員會/勞工安全衛生研究所，勞工靜態與動態人體計測資料庫之量測，行政院勞工委員會勞工安全衛生研究所，1996。
[18] 呂怡廷，“多目標基因演算法於鋼筋混凝土結構設計之應用”，碩士論文，國立交通大學，土木工程研究所，2008。
[19] 呂廣英， “簡介模組化設計與應用” ，機械月刊， pp. 121-125，1985。
[20] 李綺思，“結合基因演算法及類神經網路自動搜尋沸水式反應器升載路徑之研究”，博士論文，國立清華大學，工程與系統科學研究所，2007。
[21] 徐業良，工程最佳化設計，最佳化設計之數學模型，pp. 9-28，華泰，1995。
[22] 郭信川、王鴻鈞，“遺傳演算法在最佳化設計問題之應用”，中國造船暨輪機工程學刊，第十七卷第一期，pp. 39-48，1998。
[23] 郭啟洲， “電腦輔助三維自動化管路佈置之研究”，碩士論文，國立成功大學，造船暨船舶機械工程研究所，1999。
[24] 劉惟信，機械最佳化設計 第二版，第一章 概述，PP.1-2~1-17，第二章 最佳化設計中目標函數的數學分析基礎， PP.2-2~2-37，全華，1996。
[25] 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則”，全華科技圖書股份有限公司，台北，第9.1-9.12 頁。