- Education And Career
- Companies & Markets
- Gadgets & Technology
- After Hours
- Banking & Finance
- Energy & Infra
- Case Study
- Web Exclusive
- Property Review
- Digital India
- Work Life Balance
- Test category by sumit
Ganit Develops 'Highly Effective' Daily Forecasting Framework with Amazon Web Services
For the upcoming festive season, Ganit Inc. is working with its clients to develop bundle offers at various price segments and personalized offers
Photo Credit :
Data consulting company, Ganit Inc, which recently collaborated with Amazon Web Services (AWS) to develop a forecasting framework, claims that accuracy has increased by three times and as a result, considerable wastage has been reduced. As a consequence of the forecasting model wastage has drastically been reduced, it said.
At the same time, stock-outs have considerably reduced by 30 to 40 per cent, it added. And how did that happen? Because of day-to-day tracking model, said Shivprasad KT, Founder & CEO, Ganit Inc. He said, "Typically, forecasting models work on a weekly or a monthly tracking cycles. Now, instead of monthly forecasting, we are forecasting 365 times a year. The complexity is very high.”
He termed it as 'one of the best projects' he did in India. "AWS team was surprised in terms of the accuracy we got,” he said.
Reportedly, Ganit Inc. is the preferred AI partner for food and grocery retailer 'More Retail' (formerly Aditya Birla Retail Limited). One of the main projects which Ganit Inc. undertook while working with 'More Retail' was to understand demand pattern at the store level. Given they have over 600 stores across India and close to a hundred items in foods and vegetables in every store – the scale of complexity needed to analyze to build a solid forecasting framework was huge. Not to mention, because by the nature of it, food and vegetables have a very low shelf-life. The challenge was to find a balance between stock-out and wastage.
Shivaprasad explained: Two things happen from a business perspective. When a customer walks in the store, maybe either there is no product for the customer to buy (stock-out). Or, if you keep too many items, there is wastage. Shivaprasad said, “Only very advanced AI (artificial intelligence) models can help you achieve that. And, that’s what we achieved here. AWS team was surprised in terms of accuracy.” “Because of this customer is happier to come and shop,” Shivaprasad said.
Upcoming Festive Season Possibilities
With the festive season knocking at the door, speculations are being made on what would be this year’s major consumer trends and observations given the coronavirus pandemic have plagued the country’s economy.
Shivaprasad shared his views: “Customers are willing to pay higher prices as long as the quality of the product is maintained – that has been established post-Covid-19. Also, more space for the essentials is being allocated than the non-essentials.” The company informed that it is working with its clients in preparation for the festive season for two primary festival specific analytics. “First specific bundle offers with varied price segments are being developed. And, work is being done to make the offers more personalized.” Moreover, many people who earlier were shopping on-foot to the stores will be looking for an e-commerce option; so that has to be taken into consideration.