I think I could see a rather high value for something like this. Most inventory control schemes are very dependent on people doing something. A system like this can identify some costly errors - like when an operator uses a reel of 50v cap to cover for shortage of 35v caps - but fails to log them. The next job needs the 50v parts and they show up just fine in the 'system' - yet they will be short with no options available.
Whatever the errors - physically counting parts is obviously necessary and generally slow. In a large operation, just buying loads of excess adds up really fast. Hell, in my micro-operation, all of my excess is nearly out of control. Out of fear that I would run out, I simply purchased more stuff - now I have a lot of parts with unknown value until I look into each one. Probably talking about 300-500 unique parts and many thousands of $$. Many of them are unlikely to ever get used. Multiply that time about 100,000x in a real production operation and it would be rather crazy.