Demand estimating model to forecast the building material requirements for the construction and allied industries in Sri Lanka

Abstract

Over the past few years, there has been a high level of interest in modelling demand estimation for the construction and allied industries in Sri Lanka. Demand estimation is a process that involves coming up with an estimate of the level of demand for a product or service and, typically confined to a particular period of time, a month, quarter or year. Demand estimation methods can be categorized into two main categories according to the technique applied to analyse data. Different approaches are survey methods and statistical methods. For a good quantitative analysis, statistical methods are more preferable. Regression analysis method which comes under econometric statistical method is more preferable to develop demand estimation models since it has high accuracy level. In the regression analysis, there are two methods to develop the model. These are Single Regression Analysis and Multiple Regression Analysis. The few steps to develop the estimation models are statement of a theory or hypothesis, model specification, data collection, estimation of parameters, checking goodness of it, hypothesis testing and forecasting. In this project, there are two models for sand and ABC materials each and the first model is for dust and ¾ particle size. Developed model for chip particles was rejected due to inaccuracy of the unavailability of required data. In validation, sand and ¾ particle size have shown very high accuracy when as dust and ABC has shown quite lower accuracy.