Fuzzy and grey forecasting techniques and their applications in production systems

Forecasting is an important part of decision making as many of our decisions are based on predictions of future unknown events. Forecast is an interesting research topic that has received attention from many researchers in the past several decades. Forecasting has many application areas including but not limited to stock markets, futures markets, enrollments of a school, demand of a product and/or service. Management needs to reduce the risks associated with decision-making, which can be done by anticipating the future more clearly. Accurate forecasts are therefore essential for risk reduction. Forecasting provides critical inputs to various manufacturing-related processes, such as production planning, inventory management, capital budgeting, purchasing, work-force scheduling, resource allocation and other important parts of the production system operation. Accurate forecasts are crucial for successful manufacturing and can lead to considerable savings when implemented efficiently. Forecasting literature contains a large variety of techniques from simple regression to complex metaheuristics such as neural networks and genetic algorithms. Fuzzy set theory is also another useful tool to increase forecast efficiency and effectiveness. This chapter summarizes and classifies forecasting techniques based on crisp logic, fuzzy logic and the grey theory. The chapter also presents numerical examples of fuzzy simple linear regression and grey forecasting methodology.