Demand forecasting for LPG station by Time Series and Artificial Neural Network

Main Article Content

ไววิทย์ พานิชอัศดร
มหศักดิ์ เกตุฉ่ำ


- Demand forecasting plays an important role in business. A good inventory management can increase the competitive advantage. An inventory should be optimized so that firm can increase free cash flow and keep the company’s profitability at the same time. Especially, demand forecasting for the LPG station is very difficult because there are many factors which may not be linked directly or indirectly and it also occur in duplicate or random sometime. Thus, only the historical sale data is insufficient for develop an efficient forecast model. With the neural network technique, this technique gives an advantage in extracting the real factor which impact to demand in the future. This paper uses the artificial neural network for developing the forecast model. The result shows the presented methodology has 683 RMSE a day.

Article Details

How to Cite
พานิชอัศดร ไ. and เกตุฉ่ำ ม., “Demand forecasting for LPG station by Time Series and Artificial Neural Network”, JIST, vol. 7, no. 1, pp. 42–49, Jun. 2017.
Research Article: Soft Computing (Detail in Scope of Journal)


1. MD Zakaria Rahman, “Forecasting the Long Term Energy Demand of Bangladesh using SPSS 2011-2040”, 3rd International Conference on Electrical Engineering and Information Communication Technology (ICEEICT), March 2016. pp. 1-5

2. Iwan Awaludin, “A Prediction of Oil Demand in Malaysia Using Time Series Modeling approach”, 2008 International Symposium on Information Technology, Vol. 3, September 2008. pp. 1-4

3. Minjie Zhu, “Research of Product Oil Demand Forecast Based on the Combination Forecast Method”, 2010 International Conference on E-Business and E-Government, May 2010. pp. 796-798

4. Chengli Zhang, “Analysis and Forecasting of Oil Demand in China”, 2010 International Conference on E-Business and E-Government, September 2010. pp. 5078-5081

5. Seyedeh Mahya Seyedan, “Performance analysis and forecasting on Crude Oil: Novel Support Vector Regression application to market demand”, 2015 International Conference on Soft Computing Techniques and Implementations (ICSCTI), vol. 3, June 2016. pp. 126-130

6. Mustafa Akpinar and Nejat Yumusak, “Forecasting Household Natural Gas Consumption with ARIMA Model: A Case Study of Removing Cycle”, 2013 7th International Conference on Application of Information and Communication Technologies, January 2014. pp. 1-6

7. Yi-Hui Liang, “Using the combined model for forecasting the tourism demand”, 2016 International Conference on Machine Learning and Cybernetics (ICMLC), vol. 2, March 2017. pp. 612-615

8. Xiaoxiao Guo, "A Prediction-Based Inventory Optimization Using Data Mining Models", 2014 Seventh International Joint Conference on Computational Sciences and Optimization, October 2014. pp. 611-615

9. Thararat Udomchaibanjerd, “NAV Forecasting Model Using Data Mining and Artificial Neural Network”, The Tenth National Conference on Computing and Information Technology 2014, May 2014. pp. 890-895