Demand Forecasting using Python

A Project done for TinyCo Retail Store

Introduction

Supply Chain as a business function has almost become a metric of how well an entity is going to perform internally (in terms of meeting its own targets) and externally (wrestling against the business environment and its competitors).

According to a survey by Deloitte from 2014, 79% of companies with high-performing supply chains achieve revenue growth superior to the average within their industries. Conversely, just 8% of businesses with less capable supply chains report above-average growth. That figure highlights like no other how critical the interrelations are between an enterprise and its supply chain.

A supply chain consists of many factors which may or may not include supplier contracts, transport logistics, storage capacities and product sales. Long term contracts with suppliers often involve forecasts so that the suppliers know when to supply a company with products and in what quantities. Globalisation has recently disrupted businesses world wide. Companies now source raw materials and products from different suppliers internationally. These orders may involve very long lead times given that sea freight is the cheapest mode of transportation. To maximise on sales, a business would need to know exactly when to trigger these international orders such that they're not left with empty warehouses when the time comes. Stock orders also need to be optimised since storage space is finite.

Now the question is, how do we know when to make the appropriate triggers and what strategy do we use to make forecasts? The answer to this is Supply Chain Planning. Supply chain planning (SCP) is the component of supply chain management (SCM) that develops a strategy for balancing supply and demand, predicting future requirements and monitoring fulfilment. SCP covers a range of supply chain processes such as manufacturing, production scheduling, predictive modelling  and  sustainability.