Document Type



This item is available under a Creative Commons License for non-commercial use only


1.2 COMPUTER AND INFORMATION SCIENCE, Computer Sciences, Information Science

Publication Details

A dissertation submitted in partial fulfilment of the requirements of Dublin Institute of Technology for the degree of M.Sc. in Computing (Data Analytics) 2018.


This research project identifies the significant factors that affects the number of customer visits to a fast-casual restaurant every hour and proceeds to develop several machine learning models to forecast customer visits. The core value proposition of fast-casual restaurants is quality food delivered at speed which means they have to prepare meals in advance of customers visit but the problem with this approach is in forecasting future demand, under estimating demand could lead to inadequate meal preparation which would leave customers unsatisfied while over estimation of demand could lead to wastage especially with restaurants having to comply with food safety regulations whereby heated food not consumed within 90 minutes has to be discarded. Hourly forecasting of demand as opposed to monthly or even daily forecasting is important to help the manager of the fast-casual restaurant optimize resources and reduce wastage. Approaches to forecasting demand can be broadly categorized into qualitative and quantitative methods. Quantitative methods can be further divided into time series and regression-based methods. The regression-based approach which is used for this study enabled the researcher to gather data on several factors hypothesized to have an impact on the number of customer visits to the fast-casual restaurant every hour, carry out an experiment to test for the significance of these factors and to develop several predictive machine learning models capable of predicting the number of customer visits every hour. The results of the experiments carried out shows that hour, day, public holidays, temperature, humidity, rain and windspeed are significant factors in predicting the number of hourly customer visits. Multiple linear regression, regression tree, random forest and gradient boosting machine learning algorithms were also trained to predict the number of customer visits with the Gradient boosting algorithm achieving the lowest Mean Absolute Percentage Error(MAPE) of 18.82%.