本文提出之線性近似法克服了Miller[1991]無法找到敏感性分析資訊及 未對不同Nash均衡點重新計算敏感性分析資訊的問題，成功地構建跟隨者對於領 導者之決策行為的線性近似反應函數。文中並且以複雜之Stackelberg競局一路網 設計模式為例，與MINOS、H-J、IOA、EDO、GSDA等方法比較，實際測試線性 近似法之效能及效率。在七個測試案例中，線性近似法在效能上與先前最具效能 之方法相當，在效率上則明顯優於先前最具效率之方法。且線性近似法何用在大 型實際路網上，將來可朝實証研究繼續發展

　　　　　This paper propose a Linear Approximation Algorithm which uses sensitivity analysis information to obtain the derivatives of the reaction function and prevents the shortages of Miller's algorithm [1991]. In the article, we apply our method to two examples. The results are put into comparison with results of MINOS, H-J, IOA, EDO and GSDA. In effectiveness, our method is as well as the most effective algorithm in the other algorithms. In efficiency, LAA is superior to others. In future, we can extent our method to real network and develop quadraticreaction function.