How to prepare a process for an automation conversation?

The quality of an automation offer depends above all on the quality of the data you provide at the start. An integrator who hears "we would like to put a robot here" can only respond with generalities. The same integrator, given concrete process data, will calculate the throughput, select a solution and quote a realistic cost.

If you are still weighing up whether automation makes sense at all, start with the article when does production automation pay off. This post assumes the direction has already been decided and shows how to prepare the process so that the conversation with a supplier is concrete.

Measure the takt time and volume

Time is the foundation of every analysis. Without the takt time you cannot calculate whether automation will deliver results, or how fast the solution has to run.

• Time several full cycles and note the typical and extreme values,
• state the target takt time: how many pieces per hour or per shift you want to achieve,
• define the volume: daily, weekly, yearly,
• flag any seasonality and the expected production life of the product.

The gap between the current and the target takt time is the most important number in the entire conversation. It shows how big a performance jump the implementation is expected to deliver and whether it is achievable at all without changing the technology.

Write down the process step by step

Automation reproduces the real process, so you first have to describe it as it actually runs, not as it should run in theory.

• List the operator's tasks in order, with the approximate duration of each,
• mark the points where the operator makes a decision or inspects the part,
• flag the tasks that are strenuous, repetitive or physically demanding,
• attach photos or a short video of the station.

The easiest way is to stand at the station and note what happens to the part from entry to exit. A short recording often reveals more than a long description, because it exposes actions nobody notices any more.

Describe the part, not just the task

For automation, what matters is not only what the operator does but also what they hold in their hand. The characteristics of the part drive the choice of gripper, feeder and feeding method.

The more that is known about the part, the fewer assumptions the integrator has to make — and every assumption is a potential cost or risk during implementation.

Gather data on quality and errors

Automation makes sense when it targets a real problem. That is why it pays to know where and why errors occur before any solution is proposed.

• Which errors occur most often and at which stage,
• whether they result from mistakes, fatigue or the process itself,
• what the inspection requirements are and whether measurements must be documented,
• what a single defect and its detection cost.

If errors arise in one place, automating exactly that fragment can bring a quick result without rebuilding the whole station.

Map the part flow and the constraints

A station does not operate in a vacuum. You need to show how parts reach it and what happens to them next, as well as what the hard constraints are.

• How parts enter and leave the station: loose, in containers, on a pallet,
• where the buffers are and how transport between operations works,
• how much space there is: layout, available floor area, height,
• which utilities are available: power, compressed air, network,
• which existing machines the solution has to integrate with,
• what the occupational health and safety requirements are.

Often it is the flow, not the task itself, that is the bottleneck. If parts wait in a queue, consider a conveyor system before automating a single operation.

State the goal, not a ready-made solution

The most common mistake in an enquiry is describing a solution instead of a problem. "We want a six-axis robot" narrows the field before anyone has calculated whether it is the best route.

A better brief describes the goal in numbers and facts:

• Weak brief — "We want to automate packing with a robot",
• Good brief — "We want to cut the packing takt time from X to Y, eliminate piece-count errors and relieve the operator of lifting heavy packages".

The second description makes it possible to match the solution to the goal — sometimes it will be a robot, sometimes a simpler feeder, a positioner or a dedicated station. Which option is better is discussed in the article special-purpose machine or off-the-shelf station.

Decision framework: before you book the meeting

A quick readiness check before an automation conversation:

Summary

A good automation conversation starts with data, not with a device catalogue. Takt time, volume, task descriptions, part features, quality and flow tell the integrator more than the name of a dream machine. The more concrete the picture of the process, the more accurate and cheaper the implementation.

Want to assess what can be improved in a specific process? Contact Nomatec — we will analyse the takt time, part flow and variants, and then match the scale: from simple tooling, through conveyor systems, to full automation and special-purpose machines.