考量性別偏好限制下最佳化酒後代駕排程及指派規劃之研究

Optimal Routing and Scheduling of Designated Drivers for Intoxicated Drivers with Gender Preference Considerations

近年來，酒後代駕產業因政府推廣而快速成長，促使業者越加重視其服務效率。目前台灣的代駕司機員在實務上多自行前往執行任務，然就訪談及國外案例等資料，本研究預期未來將有業者改利用專車接送代駕司機員至現場進行服務。因此，本研究運用時空網路技巧與數學規劃方法，基於業者的角度，考量實務的營運限制及女性顧客可能要求指派同性別代駕司機員的情況，以最小化系統總成本為目標，構建代駕司機員調度排程規劃模式。此模式為整數多商品網路流動問題，可利用CPLEX軟體求解，最後再採用流量分解法，以求得各調度車輛及代駕司機員之排程。本研究以隨機生成之資料進行範例測試及敏感度分析，結果顯示該模式效果甚佳，可提供業者未來營運之參考。

In recent years, the Driving While Intoxicated (DWI) designated driver service industry has rapidly expanded, driven by government initiatives and increasing demand for efficient services. In Taiwan, designated drivers typically travel independently to their assignments. However, based on insights from interviews and international case studies, this research anticipates that some service providers may adopt a centralized model, using company vehicles to transport drivers to their assignments. This study develops a Designated Driver Routing and Scheduling Model using time-space network techniques and mathematical programming to minimize total system costs. The model is designed from the perspective of service providers and incorporates practical constraints, such as accommodating requests for same-gender drivers from female customers. The model, formulated as an integer multicommodity network flow problem, was solved using CPLEX software, with a flow decomposition method applied to determine optimized routes and schedules for vehicles and drivers. Case studies based on simulated data and sensitivity analyses reveal the model’s effectiveness in optimizing operations. These findings demonstrate the potential of the proposed approach to enhance efficiency in future designated driver services.