Portfolio Practical Assessment

Introduction

Every industry is racing to be on top to attract as many customers as they can in their specified field of domain and in order to achieve that many companies collect humongous amount of data related to their customers. It may range from personal details of the customer being on the one end of the spectrum to the high-end insights on the customer interests and spending behaviors etc.

Background

I am working as a AWS Cloud/Python Developer for a top-rated fast food restaurant industry in United States called “Chick-Fil-A” and as part of my job, I get the opportunity to look at and also work on data a little bit now and then. At one instance in my past 3 years of experience working for the company, I had come across a task where I should develop a process to transfer small amount of data for an application use (Chen, Williams & Xu,2007, May).

From the previous year’s data, it was found that Chick-Fil-A is losing few million dollars just on waffle fries on overall 2000+ restaurants. And that is when we decided to get into Forecasting of the inventory so the team started building a neural network called “Long and Short-Term Memory” to analyze and predict the inventory needed for next 3 days, predict dollar price and wastage for all the 2000+ Chick-Fil-A stores across United States. As part of building this project, team collects a lot of data about the sales transactions at a particular location, particular time and day, if the sale is drive thru or in store transaction, what is product sold, festive holidays, federal holidays sales, sales on foot ball matches at nearby stadiums, sales on a particular season, using IOT devices to predict the number of customers in a restaurant at a point of time that will let the kitchen get ready to serve the customer as soon as possible without a wait time. So the whole point of the application is to take the data points collected for each restaurant and predict the Inventory and wastage.

From all the data collected and predicted, after proper mining techniques and the outputs of the predictions show that people in Utah drink a lot of Tea instead of Coffee, so there is no point of Chick-Fil-A restaurants in Utah selling Coffee a lot thus wasting a lot of inventory in Utah for coffee. Also, stores that have Starbucks right beside a Chick-Fil-A have very low sales on Coffee, stores near to a stadium have high sales during any football/baseball games. Sales of Coffee are more in Winter and sales of milkshakes are high in summer etc. These are some of the outcomes of mining the sales data that can be of a great help to control wastage of food and inventory and also focus on marketing to boost sales during a game or a particular season.

In the same way, there are many insights that were drawn from the data collected and that information is being used to the growth of the company, putting the brand at the top of the fast food industry (Titherington,2018).

Reference

1. Chen, X., Ye, Y., Williams, G., & Xu, X. (2007, May). A survey of open source data mining systems. In Pacific-Asia Conference on Knowledge Discovery and Data Mining (pp. 3-14). Springer, Berlin, Heidelberg.
2. Titherington, T. (2018). My Head in the Clouds (Computing): A Case Study of a Restaurant Group Embracing Off-Site Technology. Boston Hospitality Review, 6(3).

Cite this page