本研究利用多項羅機運具選擇模式探討更新方法、效用函數指定、與數據型態對模式更新的影響。所探討的更新方法有貝氏更新法、轉移凡度法、結合轉移估計法、及聯合內容估計法等四種。對這四種更新方法以不同的更新樣本數來更新模式，以求出良好更新模式所需之最小樣本數。效用函數指定除一般常用的線性模式外，另建立旅行服務水準變數為非線性與旅行時間變數之參數鳥常態分配之效用函數，來探討效用函數指定對模式更新之影響。在數據型態方面則探討敘述偏好數據從事棋式更新的潛力以及選擇集合對模式更新的影響。

In this research we used multinomial logit mode choice model to discuss the effects of updating method, utility function specification, and data type on the updating of choice model. The updating method included Bayesian updating, transfer scaling, combined transfer estimation, and joint context estimation. Different sample sizes were used in these methods to find the minimum sample size for each method. Linear” utility function, non-linear utility function, and randomly distributed parameter utility function were all specified to find the effect of utility function specification on the updating of model. The potential of stated preference data for model updating and the effect of choice set on model updating were also discussed.