TEAM 13 Our team wanted to create a Java program to help predict sales within the KwikTrip business chain. The solution we found was to analyze the items that get sold at cash registers and create a program that would graph the data into a visual representation of monthly purchasing trends to predict periodic spikes in sales for certain month. To do this, we wrote a three-step process for our code to follow: find sales data in a text file, sort the file by date, and then present a visualization of the data for the user to view. Randomly generated data is used to show the code capabilities. The code to generate example data creates data from predetermined items and prices but adds in random quantities of the product, times, and dates. For example, it will give the output of a random date and time, which is constrained to the past two years, as well as the name of the item(s) purchased, the price of the item(s) purchased, the quantity of items purchased, and the location of the item(s) purchased. The next step is sorting the data - it does this by sorting the transactions involving an item by time, which allows for the easy creation of a sales vs. time graph. Then, our GUI gives the user the option to choose which text file to pull from to analyze the data. This allows them to possibly flip between two years of transactions or possibly two locations of transactions. It then sorts the data by item and loads a graph which represents the number of those items bought over the months of the year.
APPENDIX Future Improvements • Sort by Year • Predictive Modeling • Collective Database for certain Regions • Sort Item Purchases by time of day • Sort if there was a sale that day RoadBlocks • Project Structure • New Libraries (JFreeChart Library)
Log in or sign up for Devpost to join the conversation.