Paper Title
Optimizing & Improving ABB’s Global Forecast using Forecasting & Analytics Engine (A Best Fit of Statistical & Deep Learning Neural Network Extrapolated Models)

Abstract
Forecasting is a complex task to predict the future demand. It is a technique that uses historical data as inputs to make informed estimates that are predictive in determining the direction of future trends. It is an important and common data science task in organizations today. Having prior knowledge of any event can help a company tremendously in the formulation of its goals, policies, and planning. However, producing high-quality and reliable forecasts comes with challenges of its own. Forecasting is a complex phenomenon both for humans and for machines. It also requires very experienced time series analysts which as a matter of fact are quite rare.Time series modeling technique using neural networks provides a promising alternative than traditional models, businesses can manually tweak the forecast numbers to the best fit.Forecasting globally at the Country, Hub or Business unit level at scale, requires comparative study of the performances of neural network time series models for forecasting failures and reliability in forward forecasts.Forecasting is one of the most crucial parts of a supply chain and there was no single tool for part level forecasting.The users or demand planners had to glance through multiple files and screens for forecasting at part or consolidated level. And hence we thought the need of having a tool which can efficiently do the forecasts at scale. We designed a Forecasting and analytics tool to generate aggregate forecast at part level. This might be an excellent tool for management to see multiple supply chain KPIs, current inventory, forecast accuracy etc.Reliability testing of the forecast models showed that the proposed results are better performing than traditional models.This can be posed as regression problem and we can use time series to forecast the inventory in advance to maintain supply chain, this will also reduce lot of inventory cost. A solution will also be proposed to identify where to reduce the inventory to the optimum level based on the demand forecast using the machine learning algorithms.