Creating a Model to Predict Installation Agreement Revenue

Motorola Solutions is a Chicago based company that delivers mission critical products and services to various organizations. These organizations include police departments, government agencies, hospitals and paramedics, and emergency responders for natural disasters. When an organization decides to buy a product, installing that product may be necessary. This is fulfilled by buying an installation agreement. These agreements are to install equipment initially but also may take years to fulfill based on how complex the agreement may be. Because of the variation in fulfillment time, forecasting revenue from installation agreements is difficult. This project is intended to come up with an approach to forecasting the installation agreements using a variety of data science techniques. Data cleaning will be the first step in the process and may require textual disambiguation in order to extract data from comment sections. Once the data is clean, a model will have to be created using regression methods. The model will then have to be validated across a test data set. Once the model is tested, it can then be used to forecast installation agreement revenue. The objectives of this project are to use a regression method in R to model a forecast for installation agreements, to establish a reliable model that can predict revenue for the next 12 months, and to clean the data set for future use by reporting and data science teams.