Hi, I’m Srinivas and working as a Business Analyst. In my corporate, I’m responsible for creating Statistical forecasts for 450+ SKUs across 14+ countries, Ad hoc analysis of KPI evolution, scorecards on KPIs, and Financial Forecasting Analysis. I got motivated to learn data science as the concepts taught have the potential to solve a lot of business problems faced at my workplace.
The problem statement I worked on was identifying the potential risk associated with the financial forecasting submitted by the country’s finance leads. To diagnose the problem, I started by applying concepts of the Central Limit Theorem and Confidence Interval in Creating Range Forecast at the annual level which in turn helped in identifying the Revenue at risk for 30% of the top Portfolio. The activity proved to be beneficial as it helped in avoiding risks associated with capacity planning that includes the purchase of a new manufacturing plant and working capital being locked in the inventory.
In this exercise, I utilized Python for Monte Carlo Simulations and Interval Forecasting. The tools were selected on the basis of Forecasting and Statistical Modelling capabilities and the dynamicity of the Visualization Libraries. The steps implemented for solving the problem are as follows:
- 1. Data Collection from Internal and External Sources.
- 2. Data Pre-processing for Outliers corrections and Adjustment for Actuals
- 3. Statistical Modelling
- 4. Visualizations
- 5. Results Reporting
This exercise also helped me generate useful insights. I was able to identify 50 million dollars of risk in revenue. Currently, the analysis is being done only for the top 50% of the brand portfolio, out of which 40% of Revenue is under risk. 15% of the risk is validated with proper events leaving the other 25% as unexplainable risk. I recommended the organisation scale up the analysis for 70% of the portfolio and to create a live dashboard that can help identify risks on a monthly basis. This helped us avoid extra capacity which included setting up a new manufacturing plant and working capital locked in the inventory which in turn improved the profit. I was able to increase the profit margin from 12%-16% to the top 30% of brands generally. The analysis helped in saving around 3%- 5% of the profit margin.
This exercise proved to be really beneficial and helped in learning how to employ a structured approach to problem-solving.
