‘Artificial Intelligence can account for more factors and complicated nonlinear and correlated dependencies of data much better than humans. They can do a proper risk assessment and find optimal decisions, leading to improvement in decision making.’
This is what Michael Feindt – Founder & Chief Scientific Officer of Blue Yonder, Inventor of NeuroBayes® Algorithm at CERN and Professor at the Karlsruhe Institute of Technology – firmly believes in. Ironically it was a bad decision with his own finances, leading to losses at the equity exchange, which led Feindt – a pure physicist –to realise the full potential of Artificial Intelligence (AI) and Machine Learning (ML).
“After I lost the money, I started to analyse in my free time whether one can predict equity prices with the help of an algorithm called NeuroBayes® that I wrote for CERN. This neural network can predict complete probability distributions for single observations, and for many observations. It can also be individualised, because for each observation, there are many, many input variables. I did this for physics, but then I realised, I can do this for other sectors too apart from physics,” says Feindt.
Starting Blue Yonder
Feindt started working towards his dream in 2001. Scientific work incorporating the largest data quantities in the world motivated him to find out if he could bring the same added value to business–with impressive results. In 2008, he founded Blue Yonder, a leading provider of AI and ML solutions that enable retailers, consumer products and other companies to intelligently transform their operations and make more profitable, automated business decisions that deliver higher profits and optimized customer experiences.
Before the company started specializing in retail, they worked in various sectors. The company worked with about 100 PhD data scientists –almost half of them Feindt’s students and the other half from renowned institutes including NASA. It was acquired by JDA Software in 2018. Today, Blue Yonder, a JDA company, is one of the largest and most advanced private data science companies in Europe. The company has received numerous awards, including the Experton Big Data Leader Award 2016, the BT Retail Week Technology Award and the IGD Award 2017 for Supply Chain Innovation. “Blue Yonder is based in Karlsruhe, Germany and along with JDA Software, we serve all the geographies around the world.
Our data science practice is constantly growing and will impact all of our customers and solution areas. We are focused on all major verticals–retail, manufacturing and distribution–with deep, real world experience solving some of the world’s toughest supply chain and retail problems. In India, we have our offices in Bangalore and Hyderabad,” says Feindt.
Why Retail?
“Three years ago we decided to concentrate only on retail. The reason for this was that there is very strong competition in retail, the margins are very small, and we understood that taking the right decision can make a complete difference between a very successful retailer and the retailers that are dying,” says Feindt.
“Typically Blue Yonder scientists take all historical data, add to this local holiday and festival data, factor in the historical weather forecast and make predictions for retailers based on this data.
This is a complete probability distribution, a day-to-day prediction.
In today’s supply chains, customer demand can change in an instant due to trends, global events, weather, moves by the competition and more. To be successful, new solutions are required that run on a cognitive, real time and connected platform.
The path forward begins with next-generation solutions that can turn real-time data and insights into fast, profitable business decisions. With JDA LuminateTM, retailers can now harness the power of the cloud and cutting-edge innovation – such as SaaS, IoT, AI, advanced analytics and cross- platform integration – to drive smarter, more agile supply chain transformations for greater results,” Feindt further explains.
In India, two or three years ago, a retailer looking to implement auto replenishment systems would have set up a large demand forecasting team and spent a large amount of time and money, involving BPR consultants to set up demand management processes. Today, with JDA’s Blue Yonder-infused solutions, forecast creation and forecast management is largely zero-touch. So companies that did not invest in the previous generation technologies will have smaller change management issues when they adopt these solutions.
“While they may not be upfront consumers of AI technology, even Small and Medium Retailers in India are being touched by AI/ML technology which is helping them get the right product, in time and at the right price. Smaller technology companies in cities like Bangalore are offering smaller regional apparel retailers marketing insights driven by AI solutions to create personalized content and refining marketing campaigns, and making personalized recommendations,” states Feindt.
International Success Stories
Replenishment & Price Optimization at OTTO: In the face of growing competitive pressure in the e-commerce sector, combined with rapidly changing fashion trends, OTTO turned to AI.
The multichannel retailer’s competitive environment is characterized by low margins, high competitive pressure, as well as fast changing market conditions and evolving customer demands.
The company hired Blue Yonder to continually develop AI solutions to ensure maximum effectiveness. Blue Yonder’s AI algorithm has ensured that OTTO’s forecasts are constantly improving, giving them precise results that help them plan and control a future-oriented product range.
Thanks to Blue Yonder, OTTO has little stock left at the end of the season. On a daily basis, an up-to-date forecast is made per color and size, based on hundreds of different input variables (i.e. brand, price, online placement, stock situation, weather). This means that OTTO provides Blue Yonder with millions of data records each week. Every year, more than 5 billion forecasts are created this way. A considerable business success. As a result of Blue Yonder’s input, forecast quality has improved compared to the conventional processes and the end-of-season overhang has been reduced.
The software company has helped improved OTTO’s customer experience, increased sales, reduced stock levels and decreased returns by basing strategic decisions on data insights.
Store Replenishment at Morrisons: Headquartered in Bradford since 1899, Morrisons is a British company and one of the UK’s largest food retailers with more than 120,000 colleagues in 494 stores serving over 11 million customers every week.
One of the biggest customer satisfaction barriers that Morrisons identified was product availability. The supermarket chain had relied on traditional systems to replenish its stores – mostly through manual orders made by in-store teams. This proved time consuming, created inconsistencies between stores, and was not always accurate. To improve availability, Morrisons needed a solution that could solve its two biggest challenges: understocking and overstocking.
Morrisons identified Blue Yonder as the industry leader in AI and ML innovation, and its Demand Forecast & Replenishment solution as the ideal technology to significantly enhance Morrisons’ demand planning and replenishment capabilities.
Blue Yonder launched the deployment with a three-month pilot project in the run-up to Christmas; Morrisons’ busiest and most critical trading period. Due to its success, in the following 12 months, Blue Yonder optimised ambient and longlife replenishment among 26,000 SKUs across 130 categories in all of Morrisons’ stores.
2018 saw the introduction of short life fresh products as well as fruits and vegetables to the Blue Yonder solution, making a total of 29,000 SKUs.
A key factor in the speed of implementation was the easy integration into the Morrisons IT landscape as the solution used interfaces that were already in place.
By deploying Blue Yonder’s Demand Forecast & Replenishment technology, Morrisons can now make 430 million calculations and 13 million automatic decisions every single day – at far greater scale, and more accurately and consistently than previously. In addition, the granularity of information available is enabling Morrisons to enhance the customer experience down to store level; incorporating factors like shelf sizes and consumer buying patterns, rather than taking a ‘clustered’ approach to planning and replenishment.
The technology has already:
- Improved shelf gaps by up to 30 percent
- Reduced stock-holding in store by 2-3 days
Future Plans
In India, 7 of the top 10 retailers leverage JDA to optimize their supply chain across dimensions such as category management, assortment planning, store operations and warehouse operations and all of the top five grocery and convenience retailers in India use JDA for space and category management.
JDA recently took yet another step towards their moonshot – the Autonomous Supply ChainTM – as they opened up their end-to-end supply chain portfolio as a development platform. JDA’s AI-powered Supply Chain Management (SCM) Platform is infused with AI and ML optimization technology from Blue Yonder, will uniquely enable their customers as well as application development and technology partners to create cutting-edge applications on top of JDA’s solutions for the first time.
JDA’s India business is one of the fastest growing having grown at 100 percent year on year for past two years especially since marquee Indian corporates have started to move beyond generic ERP solutions to look to differentiate themselves through their supply chain prowess.
