Jack, Districon and AIMMS have worked together for a long time. How did this relationship develop?
Well, that’s an interesting story. About 10 years ago, HEINEKEN, one of our main clients, was working with us to do distribution studies. They used AIMMS and basically told us that we needed to start using AIMMS as well. We got to know the AIMMS team and became very enthusiastic about the technology. So we started using AIMMS to perform network studies. Then around 2015, we decided to build our own solutions for end customers using the AIMMS platform. That’s how it all started. We have worked on nurturing our relationship with AIMMS since then and we now have more than 20 AIMMS specialists on our team, working in three different locations.
You opened an office in Asia Pacific recently. What are your plans for the region?
Yes, as of January this year, we have an office in Singapore. We’re starting there step by step, but as a company we have done a lot of business in APAC, especially in the airports industry. It's a great region for both Districon and AIMMS to grow.
Speaking of growth, Districon won the AIMMS Partner of the Year Award in 2015. The Award recognized the many innovative applications you have built for clients using AIMMS. Is there a project you’re particularly proud of?
It’s difficult to select only one. There are different projects that I’m proud of. I’ll start by saying that we did a great project for warehouse slotting at Peapod. This project gave us the opportunity to expand to the U.S. and become a global player. For Peapod, the results were also very compelling.
They saw a 20% reduction in stock-outs and dramatically trimmed down the amount of time it takes to slot a new wareroom: from the 500 hours it took them in the past to just 4.
I’m also proud of the AIMMS projects we have implemented in the fashion and retail industry. We enable clients to react faster in a very dynamic omnichannel environment.
Finally, from a supply chain point of view, the Nampak project we worked on in South Africa stands out. Here we delivered a differentiating capability for the client. We developed a production planning and scheduling solution which is used for profit-based business planning. The great thing about this solution is that you can work with multiple business objectives and take the best production planning decisions to match the business objective you have at that moment. From a functional point of view, it’s a solution I’m very proud of.
What was the business case for Nampak?
To just scratch the surface, let me say that within the first couple of months, Nampak was able to add a new client order to their production plan and gained significant profit. It used to take them a full week to calculate whether a full order could be added to the production plan and assess the impact of taking on new orders. They were using Excel. With the AIMMS-based solution we built, they were able to calculate this in minutes during a meeting, using real scenarios. The project had its return on investment with this use case alone.
Do you see a lot of companies making the shift from Excel to more advanced optimization solutions?
Yes, a lot of companies still use Excel for production planning, but they quickly run into limitations. With AIMMS, we can bring these companies to the next level.
What are some of the biggest challenges you encounter while implementing optimization solutions?
Well to go back to the last question, making the change from spreadsheets to AIMMS is a big step. You have to make sure the end users of the App are well-trained and you have to help clients build trust in the system, because in the end, the planners have to be able to explain the plan provided to their colleagues.
Change management is a big challenge, are there any technical challenges you often face as well?
We often struggle with balancing the complexity of the algorithms with the speed to solution. Balancing complexity with speed is often one of the most challenging parts of a project.
Data accuracy is often a challenge as well, but this is the case with every IT project. Sometimes we see that the type of data needed to make optimal decisions is not available. So it needs to be developed from scratch. In such cases, we build data validation functionality in the tool to help clients clean up their data and make it more reliable.
What are some of the most important lessons you have learned throughout the years?
There are two important things we have learned. One is that it’s always hard to estimate the effort needed and the complexity of a project. We have learned that it’s very important to start with an in-depth process understanding before building a model. This helps to get a better grasp of the requirements and sets the best conditions for us to work in an agile way with clients.
The other big lesson is that end users really need to be on board with a project. They are the ones that need to work with the solution, so they need to understand what’s happening when they change certain parameters. This will help to gain trust in the outcomes. Incorporating them in the early stages of the project is very important.
Connect with Jack on LinkedIn to learn more about Districon. Eager to optimize global supply chains? Districon is looking for AIMMS experts in Germany and the U.S. Visit their careers page to learn more.