Scheduling of unrelated parallel manufacturing cells with limited human resources

In this paper the scheduling of batches of parts within a manufacturing system consisting of M unrelated parallel CNC manufacturing cells and employing a limited workforce capacity is investigated: the objective is to maintain the number R of available workers lower than the number M of the cells leaving unchanged the whole system output. Each cell of the system consists of a CNC machining centre which works batches of small mechanical parts, grouped together on a common pallet in accordance to the group technology principles. Batch splitting among CNC cells is not allowed and the setup time required by each batch is both sequence and cell dependent. The setup activities are fully performed by the workers employed within the manufacturing system; the working operations within the CNC cells are completely automated, thus the workers tasks are restricted only to the setup operations. The effect of reducing the human resource capacity on scheduling of parts within cells, and on several process performance measures, is investigated through an extensive design of experiment involving different process scenarios. The process activities and resources dynamics are simulated and optimized by means of an integrated simulation framework, consisting of an object-oriented simulator interfaced with a heuristic procedure, designed within the eM-Plant software environment. The obtained results show how the human resource can be maintained significantly lower than the number of cells without affecting the output of the manufacturing system.