車輛偵測器為蒐集交通資料重要的來源，當車輛偵測器資料回傳至交通控制中心可能會遇到車輛偵測器當機、故障等因素造成車輛偵測器資料遺漏或錯誤，本研究藉由資料插補的概念推估出完整可靠的交通資料。針對雪山隧道車輛偵測器資料，使用K-means將蒐集到的車輛偵測器資料進行分群，再以回饋式類神經網路插補遺漏資料，假設兩個偵測器資料插補中間任何一個位置的遺漏資料，其插補績效皆不錯，即可將遺漏資料之位置視為沒有設置車輛偵測器的必要，即可拉長車輛偵測器的佈設間距。研究結果顯示，在考量插補流量特性下，可設定佈設間距為11,900 m，此時準確度為80%，相當於目前35個車輛偵測器可以縮減成2個，減少車輛偵測器之設置，藉此降低車輛偵測器設置成本。 The main purpose of this study is to use the concept of missing value treatment to investigate the maximum installation spacing of vehicle detectors on road sections. Assuming a vehicle detector undergoes data loss, data error or transmission distortion, we supplement its traffic data with those from its up and downstream detectors.By means of performance assessment, we identify the farthest effective detectors for supplement, and, hence, conclude the maximum possible installation spacing according to the distance between them. An empirical analysis for the missing value of vehicle detectors in Hshehshan tunnel, we, clustered all the data into groups using K-means, and then chose to recurrent neural network impute the missing data.We, finally, developed two possible applications based on imputation performance, including data imputation and installation spacing of vehicle detectors.The result shows that we could extend the current spacing of 350 m to 11,900 m by an accuracy of over 80%.
