Quality-aware bus-driver scheduling for intelligent transportation system

Driver scheduling is an integral component of the Intelligent Transportation System (ITS). It improves travel efficiency by reducing traffic bottlenecks, irregularities, and accidents and enhancing passenger safety, and driver quality in public transport systems. The scheduling and maintenance of effective timetables are the biggest challenges in developing nations. Furthermore, competent and skilled drivers are not compensated extra for their work. To solve this problem, we proposed a bus transportation system that included a journey schedule and driver scheduling algorithm. Drivers are ranked based on their skill sets and standards. The schedule incorporates peak-hour passenger volume and schedules a set number of trips and lines. The drivers are selected based on their skills and attributes, encouraging them to improve and follow traffic laws, making the trip safe and secure. Our Quality-Aware Optimal Solution (QAOS) allocates the best drivers to complete journeys following labor rules. An alternative Quality-Aware Greedy Solution (QAGS) can complete the same number of trips in less time with more drivers due to the problem’s NP-hardness. Experimental results from a real-world case study reveal that our approach eliminates bus driver laborers, regulates labor limitations, maintains rest hours, and assigns qualified drivers to trips. • Introduced a noble framework for scheduling bus drivers that prioritizes qualified drivers for trip allocations. • The uniqueness of this work is its prioritization of allocating highly qualified drivers to complete all trips, hence minimizing the likelihood of roadside accidents. • The optimal set of drivers for trip allocation was determined using a Mixed-Integer Linear Programming (MILP) function while maintaining the labor regulations. • A greedy solution has been implemented to mitigate the NP-hardness of the MILP formulation by prioritizing route assignments for qualified drivers. • The performance analysis exhibited significant improvements in trip allocation and driver workload when compared to existing methods.