Free preview. This lesson is part of a paid course in the ETA Robotics Fluency. About this course
From fit check to a scoped project
In this lesson you'll
- Turn a Robot fit check into a scoped project with a problem statement and baseline
- Check whether the process is stable enough to automate
- Write a one-page project scope your team and an integrator can use
Welcome to Level 2
Welcome to Robotics Fluency: Practitioner from the Enterprise Technology Association (ETA). In Robotics Fluency: Foundations you learned the kinds of robots, physical AI, jobs and safety basics and ETA's Robot fit check: Task, Space, Safety, People and Payback. This course takes the next step: scoping a real project, choosing the robot and partner, running a pilot, integrating it, making the safety case, measuring results and scaling. It's for operations, engineering, EHS, IT and project leads who will run robot projects, not just talk about them.
- Step 1: Assess
Is this a good task for AI, and what's at stake?
- Step 2: Direct
Brief the AI clearly: goal, context, constraints, done criteria.
- Step 3: Assure
Check the output before you rely on it.
- Step 4: Protect
Keep people, data and your organization safe.
- Step 5: Tune
Iterate, then save what works for next time.
Use the steps in order on every task, and loop back to Tune when results fall short.
Start with the problem
A good scope begins with a problem, not a robot. Write it in one sentence with a number: "End-of-line palletizing on line 3 causes about a third of our recordable injuries and needs two people per shift we can't keep hired." Then capture a baseline: rate, labor hours, quality, injuries or ergonomic scores, downtime and cost, with where each number came from and over what period.
Is the process ready?
Robots are fast at doing the same thing. They are bad at surprises. Before automating, check:
- Variation: how many part types, sizes and orientations? How often do they change?
- Exceptions: jams, damaged parts, missing labels, mixed loads. Count them over at least a full week.
- Upstream and downstream: does material arrive steadily? Can the next step take the robot's output?
- Standard work: do all shifts do the task the same way?
If exceptions are frequent, fix the process first. A simple chute, a better label or a standard tray can make a robot project succeed; skipping that work is the most common reason projects stall.
The one-page scope
- Problem and baseline with sources.
- Fit check results: task, space, safety, people and payback, each with a short note.
- In scope and out of scope: which line, which products, which shifts.
- Success measures and targets: for example, 95% of planned rate, 90% uptime, zero injuries, payback under three years.
- Constraints: budget range, space, utilities, downtime windows, union or policy commitments.
- Owner and team: a named project owner and the operator, maintenance, safety and IT people involved.
Workplace example: a snack food plant
A snack plant wanted "a robot for packing." The project lead ran the fit check and wrote a scope: case packing on line 4, three box sizes, 18 cases a minute, two operators per shift, highest ergonomic score in the plant. A week of counting found 4% of bags arriving folded, which jammed the infeed. A guide fix cut that to under 1% before any robot was bought. The scope went to three integrators with the baseline data attached.
Try it
In your approved AI tool:
Goal: Turn my Robot fit check into a one-page project scope.
Context: Problem: [describe]. Baseline: [rate, labor, quality, injuries, with sources]. Fit check notes: [task, space, safety, people, payback]. Exceptions counted: [list with frequency].
Constraints: Use the six-part scope (problem and baseline, fit check, in and out of scope, measures and targets, constraints, owner and team). Flag any sign the process isn't ready and suggest fixes first. No vendor names.
Format: One page with six headings.
Done criteria: An integrator could start a quote from it.Review it with an operator and your maintenance lead.
Key takeaways
- Start with a problem statement and a sourced baseline.
- Count exceptions and fix unstable processes before automating.
- A one-page scope with targets, constraints and an owner keeps everyone aligned.
Sources
