為了順暢交通的流通，直行／右轉共用車道在停止線下游通常設有機車兩段式左轉待轉區，上游則設有機車停等區。本研究利用在臺北都會區所蒐集之現場資料，以探討機車兩段式左轉待轉區、機車停等區及機車停等區上游車道之停等車疏解特性，並建立分析模式。現場資料顯示，機車兩段式左轉待轉區之每公尺待轉機車數與疏解時間，有統計上顯著的線性關係；而其疏解時間通常很短，故很少影響停止線上游車輛之疏解。從機車停等區能疏解之停等機車數與疏解時間，可利用簡單的模式來估計。機車停等區上游車道之混合停等車輛的疏解行為相當複雜，但可以真實地以一 8×4×1 的類神經網路模式來代表。本研究整合這些分析模式，成為一可用以估計具有機車停等區之直行／右轉共用車道容量的模式。

To facilitate traffic movement, a shared straight-through/right-turn lane may have a two-stage left-turn box downstream of the stop line and a waiting box upstream of the stop line for motorcycles. This study uses field data collected in the Taipei metropolitan area to examine and model the queue discharge characteristics of the vehicles in the two-stage left-turn box, the waiting box, and the roadway upstream of the waiting box. The field data reveals that the queue discharge time in the two-stage left-turn box has a statistically significant linear relationship with the number of motorcycles per unit width of the box. The discharge time from the two-stage left-turn box is usually very short and thus rarely affects the discharge of the queuing vehicles at the stop line. The queue discharge time and the number of motorcycles that can be discharged from a waiting box can be estimated from simple models. The queue discharge characteristics of the mixed vehicles upstream of the waiting box are very complex but can be realistically represented by an 8×4×1 artificial neural network model. This study integrates these models into one to estimate the capacity of a shared straight-through/right-turn lane that has a waiting ox.