Metrics to Track in Partial Production

When designing a tracking system for production, the biggest mistake is trying to use the same metric for every operation. “Is it done?” doesn’t mean the same thing for every step. The signals you should track change with the operation. Rigid, universal systems don’t work.

Simple Binary States Aren’t Enough

Most companies limit work to two states: done or not done. This system is easy to use, but it doesn’t reflect reality. An operation actually progresses through a spectrum — not from 0% to 100%, but with many intermediate states along the way.

Today, an alternative is possible: operation count tracking. Instead of measuring whether a job is “completed / not completed,” you measure the number of steps it passes through, weighted by each step’s contribution. This yields forecasts that are tighter and closer to reality.

Example: Welding Operation and Weight Tracking

Take welding operations. In most systems this job appears as “done / not done.” But in the welding process there’s a far more meaningful signal to track: the weight of material to be welded.

This metric delivers two critical pieces of information at once:

• Relationship to project volume: Knowing how many kilograms need to be filled in a part gives you a direct way to measure remaining work. Instead of “Is it finished?” you answer “How many kg are left?”
• Workshop welding speed: Converting that same data into kg/h (kilograms per hour) reveals the workshop’s real throughput. Which machine is faster, which shift is more productive, which tool produces more — these aren’t just reports, they become the pulse of operations.

The Key to Sound Investment Decisions

With this information, investment decisions change completely. In the past, “should we buy a machine?” decisions relied mostly on intuition or simple demand forecasts. Now you can:

• Capacity planning: With real kg/h data, you know what each machine will produce.
• Machine investment: Actual throughput data directly supports whether to buy new equipment.
• Material procurement: With volume and speed forecasts, material planning becomes reliable. Risks of excess stock or supply interruptions drop.
• Shift and scheduling: Knowing a machine’s real capacity lets you set shift counts and working hours on solid ground.

Conclusion

In partial production, the metric you track defines the operation. Using measurable signals like operation count and weight instead of a simple “done / not done” binary state produces forecasts that are closer to reality.

This isn’t just reporting — it directly improves the quality of fundamental decisions: investment, capacity, and procurement. The point is for your production system to tell you not just what happened, but what will happen.