Master of Business Administration

Faculty / School

Faculty of Business Administration (FBA)


Dr. Rameez Khalid, Assistant Professor, Department of Management

Committee Member 1

Dr. Rameez Khalid, Assistant Professor, Faculty of Business Administration, Institute of Business Administration, Karachi

Project Type

MBA Research Project

Keywords">Imtiaz supermarket, Standardization, Naive approach, Exponential smoothing, Holt’s model, Winter’s model, KPIs

Abstract / Summary

Imtiaz supermarket (ISM), being a leading retail chain in Pakistan, is in need to standardize the forecasting model for its various product categories. Currently, the forecasting is done based on experience of the purchasers. The downside of this subjective approach is its lack of standardization and dependency on the experience of the purchaser. This approach frequently results in stock outs or excess inventory in warehouse. As a result of this, huge working capital can be stuck up in Inventory. The project report is an attempt to standardize the forecasting methodology based on secondary research and multiple mathematical models. Such a standard model will eliminate the subjectivity and experience issues. Moreover, by increasing forecasting accuracy, working capital requirement of ISM could also be reducing. For the purpose of this project particular product categories were selected by ISM and their monthly sales data was shared with the project team. After testing multiple models i.e. Naive approach, exponential smoothing, moving average, Holt’s model and Winter’s model, the project team selected the best fit for each category having the lowest MAPE. A provision for safety stock was also provided in the model and weekly ordering plan mechanism has also been proposed to further manage the working capital. A brief overview of retail sector’s KPIs has also been presented to set benchmark for ISM to gauge its performance against the competitors. To ascertain the accuracy of model, the actual sales of some categories was compared against forecasted numbers driven from the model and the variance was found to be within tolerance.

Available for download on Wednesday, December 02, 2026