Hybrid MTSR–AMR warehouse system with class-based storage policies: modeling and performance evaluation

This paper presents a performance analysis of a hybrid warehouse automation system integrating Multi-Tote Storage and Retrieval (MTSRs) autonomous mobile robots and Autonomous Mobile Robots (AMRs). A mathematical model is developed to estimate the system sizing, throughput and to identify bottlenecks based on robot cycle times and queuing effects at the input/output (I/O) station. Three scenarios are investigated: (i) a system using only MTSRs, (ii) a hybrid configuration combining MTSRs and AMRs, and (iii) a class-based storage layout that assigns high-rotation items to AMRs only. The results demonstrate that hybrid configurations significantly reduce the number of robots required and improve overall throughput. The best performance is achieved with the class-based approach, where the system reaches the maximum throughput with fewer MTSRs, thanks to task segmentation and intelligent item allocation. The study provides design guidelines for scalable and efficient warehouse systems in high-density, high-variety environments.