Seeing waste: the observation discipline behind the categories

Standing still long enough for the pattern to appear

Choose one operation and watch several consecutive cycles without interrupting. The first pass shows the obvious. Later passes show the variation: the cycle where the operator hunted for a spanner, the one where a component would not seat, the one where they waited for a crane. Note the deviations rather than an average, because the deviations are the improvement opportunities. Be honest about the observer effect. People work differently while being watched, so stay long enough that the novelty fades, explain plainly why you are there, and validate what you saw with the operator afterwards rather than questioning them mid-cycle.

The categories are a prompt, not a scoring sheet

Overproduction, waiting, transport, over-processing, inventory, motion, defects and unused capability exist to make you notice things you would otherwise walk past. They are useless as a filing system. A team that spends its afternoon debating whether an item is transport or motion has converted an improvement exercise into an administrative one. Record what you saw in plain language, note where and when, and leave the label off if it is contested. What matters is whether the observation is specific enough that somebody could go and confirm it, and concrete enough to suggest a change.

The waste that belongs to the people doing the walking

A walk that only examines operator movement is politically safe and generally the least productive. Much of what a floor wastes was created elsewhere: a tolerance tightened by engineering that requires a second operation, a batching rule set by planning that fills the aisles, a purchasing decision that forces sorting on receipt, a quality check added after an incident and never reviewed. Naming these requires the relevant department in the room and a manager willing to hear it. Where that is not possible, expect the exercise to find only what operators can be asked to do differently.

Separating waste from necessary non-value work

Not everything that fails to change the product can be removed. Regulatory records, customer-mandated inspection, calibration, safety isolation and movement imposed by the shape of the building all consume effort without adding value, but they cannot simply be deleted by a team on a walk. Keep two lists. One holds work that can be reduced or removed locally. The other holds work that is only necessary because of a decision somewhere else, which needs escalation, a customer conversation or a change to the building. Blurring the two produces either paralysis or reckless removal of something that was there for a reason.

What to do with a list that is too long

Most waste walks generate more findings than any area can act on, and an untriaged list demoralises people faster than doing nothing. Sort by who can act: items the team can change this week, items needing a fitter or an engineer, items needing money, items needing a decision from another function. Do the first group immediately, because visible action is what keeps people reporting. For everything rejected or deferred, say so and give the reason. Silence about the majority of a list teaches people that observation is pointless, and the next walk will find nothing.

Frequently asked questions

Should operators be trained to identify waste themselves?

They should, because they see the same operation continuously and already know most of what goes wrong. The obstacle is rarely awareness; it is that nobody has asked, or that previous reports went nowhere. Keep the training short and practical, done at the workstation on their own job rather than in a classroom with generic examples. Then make sure the first few things they raise are dealt with visibly, because that determines whether anyone bothers a second time.

How do we quantify waste without inventing figures?

Measure the specific thing you observed rather than estimating a category total. Count how many times in a shift an operator leaves the station and where they go. Record the clock time material sat between two operations over a week. Log the occasions a machine stopped and the reason. These are countable, defensible and directly connected to a change you could make. Broad estimates of what waste costs the plant tend to collapse under scrutiny and damage the credibility of everything else in the report.

Is it worth doing a waste walk in an office or planning function?

Frequently more worthwhile than another walk on the shop floor, because administrative processes are rarely observed by anyone. Look for work waiting in queues between people, information rekeyed between systems, approvals nobody reads, and reports produced for an audience that no longer exists. The observation technique is the same, though the waiting is measured in days rather than minutes and the evidence sits in inboxes and shared folders rather than on pallets.