Forecasting Rubber Production Using Intelligent Time Series Analysis to Support Decision Makers

Abstract

Decision support system (DSS) have become a significant factor for many organisations as assistive tools for managers to deal with problems (Nakmuang, 2004). Although DSS are used globally in the agricultural, public, government and especially in business sectors, they have not been effectively utilised within the public agricultural rubber industry in Thailand. They assist decision makers to complete decision procedure activities, obtain data, documents, knowledge or models, increase the number of alternatives examined, achieve better understanding of the business, provide fast responses to unexpected situations, offer capability to carry out ad hoc analysis, obtain new insights and learning, facilitate improved communication, achieve cost savings, achieve better decisions, facilitate more effective teamwork, achieve time savings and better use of data resources (Keen, 1981; Olson & Courtney, 1992; Power, 2004; Royal Thai Army, 2007). DSS also present graphical information and may be integrated with expert systems (ES) and artificial intelligence (AI) and support both individual and group decision makers (Power, 2004, 2007). Intelligent DSS demonstrate a range of capabilities and have the capacity to deal with complex data or problems. Hence, the evolution of intelligent DSS has demonstrated increasing functionality, including data mining, geographical information systems (GIS), business intelligence (BI), group DSS (GDSS) and hybrid DSS (Intelligent Science Research Group, 2002; Power, 2007). These functionalities are applied in intelligent DSS in a wide range of sectors, especially tourism, agriculture, industry and commerce (Intelligent Science Research Group, 2002; Power, 2007). Forecasting, as a significant capability of decision support systems, provides useful information and supports organizations by facilitating enhanced and desired performance or management in decision making. Moreover, forecasting is critical within industry because it enables prediction of future events and conditions by statistically analyzing and using data or information from the past (Markland & Sweigart, 1987; Tomita, 2007). Results from forecasting directly affect organizations in the areas of management, planning, production, sales and prices (Geurts, Lawrence, & Guerard, 1994; Markland & Sweigart, 1987; Olson & Courtney, 1992). Therefore forecasting requires a trustworthy tool to enhance accuracy before management decisions may be made.