Heuristics are one-page rules for operations and supply chain, from a practitioner who keeps learning and applying them.
In 2021, I decided to strengthen my intellectual muscles and began a PhD journey in Logistics and Telematics. My initial research focused on developing a framework for designing and deploying autonomous supply chain systems. A few years later, LLMs entered the picture. Several sparring sessions with ChatGPT made me rethink my approach, leading me to withdraw from my studies in 2025 and launch the Automatiqa Lab project. But the appetite for research didn't disappear with it - it just got blended with the real life of operations and supply chain strategy.
That blend produced this series of one-page heuristics. Each one contains key definitions, rules and principles, test cases, and awareness of potential mistakes. They aren't blind frameworks, but rather strategic reflections of a conscious practitioner.
Where the rule found me and what it looks like on the shop floor.
The forces that keep it unsolved - what every fix costs.
One instruction, in the imperative. What to do when the test says yes.
How to tell it applies to you, then the rule run on one real case.
The ways the rule gets misapplied - most of them ones I made.
One line to carry it out of the room.
A problem optimised is a problem kept.
Improvement programmes make issues cheaper and keep them alive. An ideality mindset actually asks what must be true for the problem not to exist.
A company is never ready. A loop is, or it isn’t.
Supply chain and operations AI readiness tends to score the wrong item. The decision loop determines whether it’s AI-ready.
Diagnose the shape before you prescribe the cure.
Symmetric vs asymmetric supply chains, and the Outside Variance Ratio that tells you which one you're in.
Match the variety or pay for the difference.
Any operation can absorb only as much variation as it has responses for. The Variety Shortfall Test finds where it loses.
Every buffer is a decision someone couldn't make in time.
Stock, spare capacity and slack are decisions taken early, because the loop cannot decide in time.
The same writing is mirrored as markdown at github.com/alxsidr/heuristics, licensed CC BY 4.0. Feed: /heuristics/rss.xml.
New writing, when it lands
No ads, just practitioner notes on supply chain and AI - what worked, what did not, and where all this brings us.
One click to unsubscribe, in every email. Your address goes to my distribution list only and not shared with any 3rd party.