How an automation buyer defines the problem before asking for a quote, and why you should write it down

The call starts with "we need a robot to load the press" or "the line is too slow, send a quotation". That is a complaint, not a specification. Quote against it and you are guessing, and so is every rival. Help the buyer define the problem first. Walk the line, time the station the buyer blames, and find what it waits for. Then send back one page with the symptom, the cause you saw, the number that proves it and the target. Ask for corrections before you price. Your quote then answers a problem the plant has already agreed.

What the buyer knows, and what is a guess

The buyer knows the symptom well. Orders are piling up, the second shift misses its count, workers leave after the festival and do not come back. The production register says so, shift by shift.

The cause and the cure are a guess. "Robot" is the word heard at an exhibition or seen in a bigger plant. Often nobody knows the cycle time at each station, how long the bottleneck stands idle, or what a changeover costs. Nobody in the plant has measured it, because nobody had a reason to.

That gap is your opening. The integrator who measures first is the one whose proposal reads as if it was written for this plant. The meeting where you then present to the plant head and the engineer together is covered in presenting to a plant head and an engineer.

Walk the line before you write a number

Book the visit for a working shift, not a meeting room. On the walk, cover these:

• The station blamed. Stand there through several full cycles, then go upstream and downstream. Does it wait for parts, or do parts wait for it?
• A stopwatch and a notebook. Time cycles, waits and one changeover. Your own timings beat anyone's memory.
• The production register. Output per shift for a few weeks, set against the order book.
• Variants. How many products run on this line, and how many changeovers a shift sees.
• Hands. Who works the station on each shift, especially the night shift and the weeks when workers go home.
• Rejects and rework at that station, from the quality register, not from recollection.
• The site. Floor space, power, compressed air, and the state of the floor where a machine would stand.
• History. What was tried before, and why it was dropped.

Ask the maintenance head one question before you leave: what would be hardest to look after? The answer shapes the design more than the plant head's wish list.

Symptom, cause and the number that proves it

Every complaint has three layers. The symptom is what the plant says: the line is slow. The cause is what you saw: the press stands idle while it waits for the next load. The number is what proves it, measured in the plant's own unit, such as minutes of idle time per load across a shift.

The rule is simple. Never price a symptom. If you cannot find a number, you have not found the cause yet, and the visit is not finished.

Set the target in the same unit as the number. "Improve productivity" cannot be checked at handover. "Press idle time per load cut to the time it takes to open and close" can be.

The one-page problem statement

Send this within a few days of the walk, while the plant still remembers your visit.

Problem statement: [line or station], [plant name]

What the plant sees: [the symptom, in the production head's words]

What we measured: [station, method, shift and date] and [the number]

Where the time goes: [the cause, with the evidence]

What must not change: [products, operators, floor area, quality checks that stay]

Target, in the same unit: [figure], checked against [the plant's own register]

Variants and changeovers to cover: [list]

Limits: [shutdown window, power, air, who holds the budget]

Questions for the plant: [list]

Please correct anything above. We will prepare the proposal once it is right.

Send it to the production head and copy the maintenance head. Their corrections are worth more than their approval. Keep your design out of this page. It describes the problem, and it may well be forwarded.

For example: a hot press in Yamunanagar

Imagine a plywood factory in Yamunanagar, Haryana, whose owner asks for "a robot to load the hot press" because output is short. The details are illustrative. The first integrator met the owner in the office and quoted a press loader sized to that description. A rival quoted lower. Neither had timed the press, so the owner compared two prices for a guess and did nothing.

The integrator went back and asked to watch a full shift. Its engineer found the press was not slow. It stood idle between loads, waiting for the layup tables, where workers sorted core veneer by thickness as they laid it. On the night shift, with fewer hands, the wait grew longer.

The problem statement said exactly that: idle time per load on both shifts, the cause at the layup tables, and a target of keeping the press loaded. The production manager corrected the number of tables and added that veneer reached them unsorted from the peeling section. The proposal that followed covered a veneer sorting station and a loader, priced against press idle time. The rival's quote, for "a robot", no longer answered the question the plant was asking. The extra shift of watching and the statement cost the integrator about a week of an engineer's time before any order, and the sorting station made the total price higher than the owner first expected, which took another round to settle.

When the answer is not automation

Sometimes the walk shows a cheaper fix: one more layup table, a changeover kit, or a supply problem upstream. Say so plainly. You may lose this order, but you are the integrator who measured.

Skip the full exercise when the scope is already fixed, such as replacing an old relay panel like for like, or when a consultant has written a detailed specification. Then question only the gaps.

Write the next problem statement before the next price

On your next automation enquiry, hold the quotation until the one-page statement has come back corrected. Count how many of those proposals then go through without a revision. Buyers search for the symptom rather than the controller, which is the theme of the industrial automation and robotics page, and writing the proposal that follows is proposal development. If you want the routine set up for your team, tell us what you need.