Human Performance of Manual Sorting: A Stochastic Analytical Model

The sorting activity is a very common task mainly executed in sectors where there is the need to sort items with specific features from the general material stream moved through a conveyor belt (e.g., waste management, agro-industrial, production lines). Despite the increasing employment of sorting vision systems that automatically select the component to be picked up, the sorting activity is still manually executed by operators. In this context, the behaviour of operators is mainly studied under an ergonomic point of view, without analysing its performance by considering the likelihood of picking up the items. Authors propose a novel stochastic analytical model that allows to evaluate the number of items that can be manually sorted in a given time window. The novelty of the model relies on modelling the likelihood of picking up an item placed on a specific location of the conveyor’s belt as a probabilistic function of the time needed to execute the entire sorting movement, based on the sum of the standard time of basic movements. Results obtained are in line with values available in the scientific literature, confirming that the novel stochastic analytical model can be employed as a tool to predict the operator’s sorting performance. Future developments will be focused on considering more complex scenarios with multiple type of items to be sorted, different handling strategies, as well as multiple operators placed on the conveyor’s belt.