A novel system dynamics model for forecasting naphtha price

摘要

Fluctuations in naphtha price are directly related to the profit of petrochemical companies. Thus, forecasting of naphtha price is becoming increasingly important. To respond to this need, a naphtha crack (the price gap between naphtha and crude oil) forecasting model is developed herein. The objective of this study was to design a reasonable forecasting model that is immediately available and can be used to develop various naphtha supply strategies. However, it is very difficult to forecast a price value with a high accuracy. Therefore, the proposed model focuses not on the price value but on the direction of the crack. These considerations are vital to a company's decision-making process. In addition, a system dynamics model that considers causal relations is proposed. It was developed based on heuristics, statistical analysis, seasonal effects, and relationships between factors that affect naphtha price, and it exhibits an accuracy rate of 84%-95% in forecasting of the naphtha crack three months in advance.