Preference-oriented crew rostering optimization model for urban railway transit

Abstract

With the continuous development of urban rail transit in China, the line operating mileage is gradually increasing, and the line condition also changes from single point attendance to multi-point attendance. In the view of multi-point attendance and multiple parking lots on urban rail transit, drivers will have some preferences, such as a preference to exit at a certain point of attendance or not wanting to make a night transfer from a depot to another. To solve this problem, we propose three preferences that affect drivers' work efficiency and happiness under the condition of multi-point attendance, including rotation preference, night-morning continuous task preference and attendance point preference, and hence construct preference-oriented crew rostering optimization model for urban railway transit. Taking Shanghai Metro as an example, the proposed model is fully tested. It is proved that the model is effective for the above problem. It also shows certain advantages compared with the traditional manual programming method in the demand of rotation leave, the allocation of night and morning break time and the demand preference allocation of attendance points, which provides research basis for urban rail transit crew rostering problem.