Green robotic warehouses: analysis of carbon emissions in a rack-climbing robotic warehouse

Our research is motivated by evaluating the CO $ _2 $ 2 emission and identifying strategies to reduce the CO $ _2 $ 2 emission in a rack-climbing robotic warehouse that handles both expedited and standard orders. We investigate the impact of both assignment policies (shared and dedicated) and priority policies (dynamic versus static) on throughput time and CO $ _2 $ 2 emission for expedited and standard orders, taking into account battery management. We propose a dynamic priority semi-open queuing network to model a dual-class order system in an e-commerce setting, incorporating the challenge that the probability of robot battery charging is not known in advance. We propose an iterative approximation analytical algorithm to solve the model. The results show that: (1) Compared with static priority policy, dynamic priority policy can satisfy the lead time requirement of both orders without increasing the CO $ _2 $ 2 emission. (2) The shared assignment policy can decrease the CO $ _2 $ 2 emission and shorten the throughput time of a robotic warehouse compared with the dedicated assignment policy. (3) We also provide a decision-making tool for warehouse managers to find the optimal dynamic priority parameters, maximising system profit while ensuring the maximum allowed lead times for both orders and taking into energy consumption.