Track: Operations Management and Operations Research

Abstract

Abstract

Forecasting is a process of predicting or estimating the future based on past experiences in the similar situation and using present data. It plays a crucial role in the first stage of every planning procedure. Most of the organizations fails due to lack of proper forecasting or because of erroneous forecasting. In this thesis, we will develop a sophisticated forecasting technique for the banking sector of Bangladesh. For this, we will first analyze some of the most relevant forecasting techniques such as Exponential Smoothing method, Holt’s Method and Winter’s Method. Then, we will analysis the applicability of advanced forecasting technique in the Bank Management system. Finally, combining forecasting and queuing system M/M/s/GD/∞/∞, we will develop an advanced Ad Hoc forecasting technique to predict the number of customers, number of entity of customer daily, weekly, monthly and yearly. We will observe that, this method will also help the manager of a bank to estimate the number of speakers or service men in order to serve them by producing reliable forecasts of the number of customers for each day so that the customers do not need to wait for long to get service as well as any banker will not pass time idly. Our method will also determine the length of time a customer is willing to wait to get service. This method will help the bank manager to minimize the total cost by determining the optimum number of bank tellers. Using real life data from a bank, we will compare the forecasts obtained from our advanced method with that of other forecasting methods and will present them graphically.