A Mixed Load Solution for the Rural School Bus Routing Problem

This work presents a mixed load algorithm to solve the School Bus Routing Problem applied to the rural area of a Brazilian city. We use a complete set of real georeferenced data containing 716 students, 23 schools, and the road network. Our goal is to minimize the total traveled distance of a heterogeneous fleet. We compare our strategy to a single load approach in order to highlight the improvements achieved by using a mixed load strategy. When we compare a mixed load to a single load strategy, we observe that: i) the mixed load strategy reduces the variable cost (total traveled distance) on approximately 20%; ii) the mixed load approach reduces the fixed cost (fleet size) from 28.2% up to 37.8%; iii) the mixed load approach increases the buses occupation in approximately 20%.