Currently, labor is allocated based on in-store activity – Traffic & historical sales data. Store managers should review labor throughout the week and flex up and down based on labor used and business need. External real-time factors are currently not included for forecasting.
EmpowerT was created to solve the problem of staffing rates at T-Mobile stores. At T-Mobile, stores over scheduling on average work 5%-6% more than scheduled. The number of employees needed at T-Mobiles daily is usually not predicted accurately. According to Time Forge, overstaffing occurs when too many employees are hired, there is no increase in productivity, and employees work too many shifts. In order to build this application my team utilized concepts from machine learning, Microsoft Axure and Excel. Excel was used to store employee data and Microsoft Azure was used to create a experiment using machine learning and we followed a guide on how basic prediction works and how this prediction will determine employee efficiency in this case.
EmpowerT is effective because it will give us the best judgement on how many employees need to work for a given day. Exponential waiting is a machine learning technique used to expand learning overtime within the current time frame. This application will increase employee efficiency and productivity. Overall, this will lower their over scheduling rate overtime.