Retail Demand Forecasting – Then and Now
Angshuman Bhattacharya
Entrepreneur in the AIML space with a successful exit. Building AIML organizations for two decades. Leader in client consulting for AI solutioning. (views expressed here are personal)
Gone are the days when heuristics ruled the world! There was one chief with all the knowledge about his shop, and would know how much he would sale for each product he carries. He knew when the demand will peak, and when he has to push the sales, through discounts. He was mostly correct. [Even if he is not correct, nobody would know – as no one accounting for lost sales. And no one dares to ask too ;)].
But now the world is much more complex – for better! You sale through many stores – and you don’t necessarily own them. They call the shots many a times too. So if you supply there more, they will put you on discount. If you supply less and go out of stock, they will replace you and fill the shelves with the competition. You lose in both sides!
So now it is ever important that you predict your demand better – for each of your product and in each of the store you operate.
How you can orient towards that?
- See more at: https://sibiaanalytics.com/blog/retail-demand-forecasting-then-and-now/
Supply Chain Analytics Consultant @ Mathnal | Advance Forecasting I Supply Network Optimization I Supply Chain Risk Analytics I Automation & Data Visualization using POWER BI
8 年Why do we need more stores ? if we need, how many and where they should be located ? What kind of stocks we need to manage in those Point of sales? How a risk management to be applied in case of stock out scenario ? similarly how a risk management to be applied in the case of over stock scenario? What are the distances to be maintained between point of sales to apply risk management in the scenarios of overstock and stock out scenarios? Operational challenges were there and will be there and the complexity in supply chain right now is gathering, handling & delivering high business volume. Higher the volume of business is the higher the data and that sometimes beyond the capacity of human mind to analyse and this can be sorted out using technology.