Industry 4.0-Driven Development of Optimization Algorithms: A Systematic Overview

Table 7

Typical characteristics of classical optimization models.

Property	Description	Trend
Size	The number of decision variables, parameters, and constraints	Increasing continuously because the vertical and horizontal integration increases the size of the manufacturing system (supply chain) to be modelled
Modularity	High level of modularity can help to solve the model efficiently	Important because the production systems are modular and have hierarchic structure; moreover, the modules may be solved independently decreasing the necessary solution time, and they can be solved parallel
Complexity	Complexity and nonlinearity may affect the solution time and precision dramatically	Increasing because only a complex and detailed model can describe a complex system appropriately
Time scale/adaptability	Optimization techniques often require a relatively long time to give the solution	Important because the computational time can be critical for short-term problems especially for real-time problems; moreover, the flexibility of manufacturing systems is an expectation
Solution quality	A measure of how far a solution is at most from an optimal solution is obtained	Important for long-term problems, but for short-term problems the computational time is more critical