EPanel 2.0: The Visualization Tool for Dynamically Optimizing Process of Optimization Problems

[[abstract]]在現代生活中，利用演算計算解決複雜的最佳化問題已是趨勢，而最佳化問題主要分為數值最佳化問題和組合最佳化問題。目前，已經有研究者開發出可以協助分析演算計算解數值最佳化問題的工具，讓使用者可以便捷的分析解題過程，然而，目前卻沒有有效的工具可以協助分析演算計算解組合最佳化問題。因此，本研究提出了一種新穎有用的工具，EPanel 2.0。EPanel 2.0是一個可以動態的顯示演算法解各類最佳化問題過程的視覺化界面。EPanel 2.0會視覺化顯示組合最佳化問題編碼長相，詳細畫出解的每個位元，幫助使用者觀察此位元是否為1或為0，並動態呈現演化計算的演算過程，協助使用者比較多個世代解的差異，讓演算法研究者可以簡單的觀察到詳細的解題過程，清楚了解演算法的優缺點，方便改進演算法。透過使用EPanel 2.0協助分析解題過程，能夠使得演化計算的開發更加得直覺、便利。[[abstract]]Metaheuristic algorithms are used to solve complex optimization problems, including numerical optimization problems and combinatorial optimization problems. Currently, many applications exist to help users learn how algorithms work on numerical optimization problems, however, there are no applications offering the same tool for combinatorial optimization problems. Combinatorial optimization problems usually have high dimensionality, with only two statuses, making them very difficult to analyze. This study therefore presents a novel and useful tool, EPanel 2.0, to display the process of metaheuristic algorithms solving not only numerical optimization problems but also combinatorial optimization problems. EPanel 2.0 is implemented in JAVA, making it executable in multiple operating systems. EPanel 2.0 has a clear user interface, provides an animated representation of the process, and also showing solutions in different ways on different situation. EPanel 2.0 allows users to easily analyze each step of the algorithm. In conclusion, EPanel 2.0 is a useful application in solving optimization problems.[[note]]碩