人工智慧方法應用於聖火傳遞路徑最佳化

聖火儀式為各項運動會上重要儀式象徵，而聖火的傳遞也詔告了和平與團結的信息，鼓舞大家共同參與體育運動盛事。本研究目的是應用地理資訊系統資料庫取得聖火傳遞路徑上的距離資訊，並運用人工智慧方法在總路徑最短的目標下，求得聖火傳遞的最佳路徑。本研究結合遺傳演算法（genetic algorithms, GA）及粒子群演算法（particle swarm optimization, PSO）提出一種新的混種演算法（hybrid algorithm, HA）以求解聖火傳遞路徑的最佳化問題，經模擬演算10個傳遞點的問題及比對分析民國99年大專運動會五個分區聖火傳遞路徑規劃問題，可驗證本研究所提出的混種演算法較傳統的蟻群演算法（ant colony optimization, ACO）、GA 及PSO 的演算效果為佳且運算時間較短。又與民國99年大專運動會五個分區聖火傳遞路徑比對結果，本研究所提出的聖火傳遞路徑較原規劃路徑總里程減少335公里（約11.68%），且可免去人員路線探勘的風險及費用，因此，此結合地理資訊系統資料庫與人工智慧方法的路徑規劃方法，可應用於聖火傳遞路徑的規劃上，使大型運動盛會的整體管理效能更加提升。

關鍵字

遺傳演算法；粒子群演算法；蟻群演算法

並列摘要

Ceremonial fire is an important symbol for an athletic game. Through the torch relay process, the atmosphere of peace and union is delivered and people are encouraged to participate in the athletic games together. Based on the data obtained from the global position system (GPS), in the study, various artificial intelligence approaches were used to solve the torch relay routing problem. The objective of the torch relay routing problem was to minimize the total distance of torch relay route. By reasonably combining genetic algorithms (GA) and particle swarm optimization (PSO), we developed a fast and easily implemented hybrid algorithm (HA) for solving the considered problem. The effectiveness and efficiency of the proposed HA were demonstrated and compared with those of standard ant colony optimization (ACO), PSO and GA by numerical results of the simulated instance with 10 spots and the real torch relay routing problems of National Intercollegiate Athletic Games in 2010. Numerical results indicate that the total distance of torch relay routes by HA was 335 km, which was 11.68% shorter than the original routes adopted by National Intercollegiate Athletic Games in 2010. It implies that the proposed HA can use the GPS information to schedule the torch relay routes, and it can reduce the cost of reconnoitering and management in an athletic game. Therefore, the proposed HA approach is an effective approach, and it can improve the efficiency for an athletic game.

並列關鍵字

genetic algorithms ； particle swarm optimization ； ant colony optimization

參考文獻

黃建松、許光麃(2007)。奧運會傳遞火炬及聖火點燃儀式意義之探討。國立臺灣體育學院體育學系系刊。7，61-68。

()。

Google Scholar

Chelouah, R.,Siarry, P.(2003).Genetic and Nelder-Mead algorithms hybridized for a more accurate global optimization of continuous multiminima functions.European Journal of Operational Research.148(2),335-348.

Google Scholar

Dorigo, M.,Stutzle, T.(2004).Ant colony optimization.Cambridge, MA:MIT Press.

Google Scholar

Fan, S.-K. S.,Zahara, E.(2007).A hybrid simplex search and particle swarm optimization for unconstrained optimization problems.European Journal of Operational Research.181(2),527-548.