Before the concerning of solution methodology used for machine loading problem, an understanding of its inbuilt complexities is required. Assume that take a simple illustration comprising of four machines, all having five tool slots, and the processing times used for carrying out the various operations of the part types are identified. All part type consists of four operations that can be found on any of the machines however the sequence of operation keeps unaltered. Various operations of part types can be performed on various machines along with unequal machining times and various tool slots. The adaptability of all machine and its ability of performing many various operations enable some operations assignments to be duplicated to produce alternative part routes. A quite large number of mixtures thus exist whether operations of part types can be allocated to the various machines whereas satisfying the system constraints. The problem's complexity increases exponentially whether other flexibilities as like tooling flexibilities or part movement flexibilities are taken along with the constraints of the system operational and configuration feasibility. The above operation that is machine assigned combination is to be evaluated utilizing three common yard-sticks as: system disturbs throughput, and largely flexible manufacturing systems consumption.
This is very complex to evaluate all probable combinations of operation machine allotment in order to realize minimum system disturb and maximum throughput. It is since this takes a large search space and also huge computational time. Heuristic solutions implemented to address that problems offer good solutions in restricted condition, conversely, the dynamism included in the process need more capable tools for tackling the problem. In this section, the neural network's application as a potential tool for resolving the Flexible manufacturing systems planning problem is presented.