One of the typical examples is the procurement process.
To produce products, the company needs to purchase materials. This process is regulated by internal regulations drawn up on the basis of current legislation.: up to a certain amount - the choice is based on the card, higher - through tender procedures.
Usually a purchasing engineer:
- collects requests from suppliers,
- generates a "selection map",
- and chooses a supplier based on an allegedly "objective" assessment.
The problem is that:
- evaluation criteria are often formal or outdated,
- points may be unjustified,
- the human factor (including corruption) influences the final choice.
Do you think the procurement engineer does not pursue personal motives with such a scheme, built back under Tsar Gorokha?
But why change it? The process seems to be going on, everything is "fine":
the office will write off all the bad things (illiquid at an inflated price) to the balance, and in the pocket of all those interested, you see, something will "fly".
How can artificial intelligence help?
Let's say a company creates an online service in which an engineer completes the purchase procedure and all received requests from suppliers are uploaded there.
A specific AI model corresponding to the solution of this problem is connected to the service via the API, which:
- analyzes the parameters of applications,
- identifies inconsistencies in the criteria,
- offers recommendations for improving the evaluation system,
- and directly indicates the optimal supplier based on key parameters (price, timing, quality, reliability, etc.).
Result:
- subjectivity and corruption are eliminated,
- procurement efficiency increases,
- the cost of production is reduced due to the optimal choice of raw materials.
This is how AI becomes not an abstract "technology of the future", but a real tool that saves money, systematically improves processes, and throws clueless and corrupt people overboard.
From a personal example.
I used to do a little a bed linen sewing business.
And I often came across the fact that companies requested commercial proposals, but in fact they were just interested in responding to their application for their fictitious selection card for reporting on the "objective" process. Key parameters such as fabric density (which directly affects quality and price) were not taken into account at all, although this is a very important selection criterion. The main thing for them was the price and type of kit.
For example, a half-bed can weigh 1 kg, or 2.5 kg (where the cotton consumption is 2.5 times more), but the comparison was formal. As a result, my offers were considered "expensive", although in fact they were of better quality, and they were not more expensive in terms of price relative to weight. This created the feeling that the choice of supplier was made for a pre-determined candidate, and objectivity in procurement was a formality.
Now the question is why and who might not be interested in AI in business processes? Information for reflection.
