Dynamic Workforce Scheduling and Relocation in Hyperconnected Parcel Logistic Hubs

Yujia Xu,Yiguo Liu,Benoit Montreuil

2024-05-08

Abstract:With the development of e-commerce during the Covid-19 pandemic, one of the major challenges for many parcel logistics companies is to design reliable and flexible scheduling algorithms to meet uncertainties of parcel arrivals as well as manpower supplies in logistic hubs, especially for those depending on workforce greatly. Currently, most labor scheduling is periodic and limited to single facility, thus the number of required workers in each hub is constrained to meet the peak demand with high variance. We approach this challenge, recognizing that not only workforce schedules but also working locations could be dynamically optimized by developing a dynamic workforce scheduling and relocation system, fed from updated data with sensors and dynamically updated hub arrival demand predictions. In this paper, we propose novel reactive scheduling heuristics to dynamically match predicted arrivals with shifts at hyperconnected parcel logistics hubs. Dynamic scheduling and allocation mechanisms are carried out dynamically during delivery periods to spatiotemporally adjust the available workforce. We also include penalty costs to keep parcels sorted in time and scheduling adjustments are made in advance to allow sufficient time for crew planning. To assess the proposed methods, we conduct comprehensive case studies based on real-world parcel logistic networks of a logistic company in China. The results show that our proposed approach can significantly outperform traditional workforce scheduling strategies in hubs with limited computation time.

Optimization and Control

What problem does this paper attempt to address?

This paper discusses the challenges faced by parcel logistics companies during the COVID-19 pandemic with the development of e-commerce, specifically how to design reliable and flexible scheduling algorithms to address the uncertainty of package arrivals in logistics centers and the supply of human resources. Currently, most labor scheduling is periodic and limited to individual facilities, thus the number of workers required at each center is constrained by meeting peak demand. The paper proposes a dynamic labor scheduling and reassignment system that utilizes sensor-updated data and dynamically updated package arrival demand forecasts in logistics centers to optimize work time and work location.

The paper introduces a novel reactive scheduling heuristic approach to real-time matching of predicted arrivals with the scheduled shifts of hyperconnected parcel logistics centers. During delivery, a dynamic execution scheduling and allocation mechanism is implemented to adjust available labor in space and time. In addition, penalty costs are introduced to ensure timely sorting of packages and to make scheduling adjustments in advance for sufficient time for personnel planning.

The proposed approach is evaluated through a comprehensive case study of a real-world package logistics network of a logistics company in China. The results show that the proposed approach significantly outperforms traditional labor scheduling strategies under limited computational time.

Keywords: physical internet, hyperconnected parcel logistics centers, dynamic labor scheduling, labor reassignment. The paper also introduces a rolling horizon approach to cope with uncertainty, solving the online scheduling problem iteratively and optimizing for different forecast scenarios.