What do dating apps and supply chain solvers have in common?
Kris Dockx - February 14
Dating apps and supply chain solvers are more alike than you think
Picture this: It's Valentine's Day. While you’re scrolling through your favorite dating app, looking for that perfect connection, there's a hidden world of algorithms at work — strikingly similar to those we use to optimize supply chains.
What if I told you that the art of matchmaking and the science of supply chain planning have more in common than you might think?
A simple dating app analogy
Imagine building a dating app. To keep things simple, let's divide users into two groups:
Date seekers: Users who are looking to go on a high number of dates.
Date offerers: Users who have a limited number of dates to offer.
In our analogy, Jamie is a date seeker hoping for 52 dates a year, while Alex is a date offerer who can only provide 15 dates annually. Each has their own set of preferences. Maybe Jamie values a great sense of humor or a shared appreciation for live music, for example.
In our supply chain solver, these preferences work like key performance indicators (KPIs) — such as transportation costs or differences in timing — determining how well supply is meeting demand.
Date-matching using smart algorithms
Blog post
The process starts by calculating a “match score” for every potential pairing between a date seeker and a date offerer. If Jamie attaches greater importance to a partner's ability to make them laugh, for instance, this factor is weighted more heavily in the score.
Once every pairing has been scored, the algorithm begins sorting the matches and assigns dates based on the highest scores. Let’s say the top match is between Jamie and Alex, a date offerer available for 15 dates a year. Because they're such a great fit, all 15 of Alex’s dates are allocated to Jamie, leaving Alex fully booked.
If Jamie still needs more dates, the algorithm continues pairing them with the next best available offer, and so on, until every date is assigned or every user's need is fulfilled.
Blog post
Building trust through transparency
Our deployment planning solver works in much the same way. It calculates compatibility scores between available stock and demand volumes, sorts these scores, and iteratively assigns supply to meet demand.
By breaking down the algorithm in a relatable way, I aim to make the solver’s decision-making process more transparent. When users can see exactly how decisions are made, it builds trust and confidence in the solver's recommendations. This is a prime example of explainable artificial intelligence (XAI), where the goal is to make smart solutions understandable and more trustworthy, while encouraging take-up.
While real-life dating is wonderfully complex, this playful analogy helps demystify the technology behind our solver. It shows that optimization need not be a mystery. It’s about making smart decisions based on clear logic.
Eager to deploy data science and machine learning? Curious about our solvers?
Kris Dockx
Senior Product Manager at OMP BE
Biography
With a master’s degree in mathematics and a PhD in chemical engineering from KU Leuven, Kris was already a published author before completing his master’s. Even today, his PhD research remains influential in supply chain circles. Kris founded successful start-ups in Leuven, Boston, Mannheim, and Basel before joining OMP in 2003. Initially a supply chain consultant, he now co-leads the solver team, driving impactful optimization projects.
