Agricultural Research Management Information System

Home
Research Summary
Search
About Us
- ARMIS
- Brochure
Contact Us
Report
User Request
Data Input
Help
- Operation Manual
  - PDF
  - Video
- Program Area & Commodity
We have reached 37600 number of research entries at this moment.

- Logout

Research Detail

Home
Research
Detail

Instituting a Forecasting Model for Purchasing Jute Bales in the Bangladesh Context: a Case Study on Sharif Jute Mills Limited

Md Al-Amin Uddin Bhuiyan*
Alumni, Faculty of Business Administration, American International University-Bangladesh (AIUB), Bangladesh

Sultanul Nahian Hasnat
Assistant Professor
Operations Management Department, Faculty of Business Administration, American International University-Bangladesh (AIUB), Bangladesh,

Abstract/ Executive Summary:

In this paper, we focus on preparing a forecasting model for the demand of jute bales to justify the use of modern forecasting concepts in the Bangladesh context. We study 54 months’ data from Sharif Jute Mills Limited to determine the core data pattern of jute bales requirements for yarn production. Based on our study, we classify the data with stationary patterns—including the minimal presence of seasonality—using regression analysis and graphical representation. Following the classification, we prepare a forecasting system for upcoming periods with Simple Exponential Smoothing Model. The model predicts that the production process will require 222.89 MT of jute bales in the upcoming month. Two different methods, (a) MAD, MSE, MAPE calculation, and (b) Control Chart, justify the accuracy of the model with acceptable results. Finally, we discuss research finding and future prospects so that Sharif Jute Mills Limited and similar companies may perform forecasting smoothly and improve the skill level of the procurement system to stay competitive in the global market.

Keywords: Industrial Engineering, Operations Management

Location: In Bangladesh

Start Date:

End Date:

Program Area: Socio-economic and Policy

Non-commodity: Modeling, Business

Objectives:

This applied study concentrates on preparing a forecasting model for jute bales under the Bangladesh business context for Sharif Jute Mills Limited. The study is aimed at preparing a forecast model mainly on procuring jute bales for yarn production after intensely examining the process flow system. It is possible to optimize the ordering time, lead-time, transportation system, seasonal effects and related costs using forecasting as the company makes a profit through their operations. In a broader context, this study can be a learning curve for the jute industry to set a base to order optimum quantity of jute bales from the domestic market and improve the procurement system performance.

Methodology:

Business forecasting helps to estimate the future demand using business data. It is important to understand the production process in detail and decide on the appropriate segment for data collection purposes. The study carefully examined the jute yarn production process in Sharif Jute Mills Limited and outlined a process flow diagram for the fabrication system.

According to Stevenson, there are two common approaches to forecasting—the qualitative approach and the quantitative approach. Qualitative methods consist of subjective inputs, which often defy specific numerical description. On the other hand, quantitative methods involve historical data projection to make a forecast. It usually evades individual biases that sometimes infect qualitative methods. Also, the data pattern is a significant factor in understanding how the time series behaved in the past. If such behavior continues in the future, the past pattern works as a guide in selecting a suitable forecasting method. After performing the forecast, accuracy and control of the forecast is a vital aspect. It is essential to include an indication of the extent to which the forecast may deviate from the value of the variable that occurs. Stevenson has also mentioned that it is vital to monitor forecast errors during periodic forecasts to determine if the errors are within reasonable bounds. If they are not, it is necessary to take corrective action.

The study principally focused on quantitative approach rather than the qualitative approach, used by the organization, to prepare the forecast. As historical data is available, the quantitative method can provide a better result in a forecasting model. Based on that, the study collected monthly jute bales usage data from a secondary source at Sharif Jute Mills Limited. Regression analysis and graphical method assisted to systematically analyze the data and identify the underlying series pattern. Based on the analysis, the study selected an appropriate time series forecasting model to prepare a forecast for upcoming periods. The forecasted data was justified using various methods to ensure the accuracy and control of forecast.

4. Data Analysis The study gathered 54 months (From January 2010 to June 2014) of information about actual jute bales used for production purpose from a secondary source. The analysis process explored the information from different viewpoints to understand the data series pattern and to find an appropriate forecasting model to perform the forecasting.

Source of Information: World Review of Business Research Vol. 8. No. 1. March 2018 Issue. Pp. 207 – 220

Article Link (Online Journal):

Funding Source:

Budget:

Total Budget (Tk.) :

Achievement/Findings:

The jute bale production is highly dependent upon the environment and less predictable than any time before. The industry leaders are constantly monitoring the demand patterns and developing forecasting models to predict the jute bale requirements. This research sensibly analyzes the yarn production process of Sharif Jute Miles Limited and pinpoints “Selection of quality jute bales” section to develop a forecasting system and consequently, optimize the jute bale purchasing system. The data analysis process follows a standard methodology and includes trend and seasonality identification procedure compared to the studied articles. This study develops a time series forecasting model and predicts the jute bale requirements in Sharif Jute Miles Limited. Two accuracy measures justified the performance of the forecasting models. This specific research finding is more beneficial for the manufacturing companies to improve their procurement system compared to the broad intentions of the reviewed articles. The research finding will assist the leaders to make acute decisions in adopting the demand patterns in production strategies. Future models can incorporate qualitative, environmental and economic data into their forecasting models to identify the changes in the factors influencing the demand pattern and develop an early alert system. Furthermore, this research will work as a platform for future experimentation on deploying a complete predictive analysis-based forecasting system for the jute product manufacturing industry.

Knowledge Management: Journal