Continuous Improvement in PCB Manufacturing

Continuous improvement in a board shop is a habit rather than a programme. The shops that get better every year are not the ones with the largest improvement department; they are the ones where small changes are made, written down and repeated, week after week, by the people who do the work. The programme, the posters and the training are all optional. What is not optional is a standard that is actually followed and a mechanism for changing it when someone finds a better way.

Why Improvement Stalls

Most improvement efforts stall for one of three reasons. The change is never written into the standard, so it disappears when the people who made it move on. The measure is never updated, so nobody can tell whether the change helped. Or the improvement is made by outsiders and imposed on the area, so the people who have to live with it have no reason to maintain it. Each of these is an organisational failure rather than a technical one.

A fourth reason appears in shops with a strong technical culture: the effort goes into the largest possible project and the small daily opportunities are left alone. A redesign of a whole process may take a year and produce a significant gain, while twenty small changes made in the same year by the people on the line may produce more. The two are not alternatives, but the small ones are cheaper and they build the habit that makes the large ones possible.

What Kaizen Actually Means

Kaizen is the practice of small, frequent, local change, and its defining feature is that it is made by the people who do the work with the support of the people who own the process. A kaizen event is a concentrated version of the same idea, a few days in which a team examines one area, proposes changes and implements what can be implemented immediately. The output of a good event is a changed workplace, not a report.

The events work best when the scope is narrow and the authority to change is present. A team asked to improve the whole assembly area will produce a list, while a team asked to reduce the time spent looking for stencils will produce a change. Narrow scope also makes the result visible within days, which is what maintains the credibility of the programme with everyone else in the shop.

Standard Work as the Foundation

Standard work is the documented best method for a task, and it is the basis of all improvement rather than its opposite. Without a standard there is no baseline to improve from and no way to hold a gain, because the next person will simply do the task the way they were taught. The standard should describe the key points and the reasons for them, so that it can be improved intelligently rather than followed blindly.

A standard should also be short and live where the work happens. A laminated sheet at the machine, a setup card on the press and a checklist on the traveller are all forms of standard work, and all of them are more useful than a manual in an office. When a change is agreed, the standard is edited the same day, because a standard that is updated a month later is a standard that will not be followed in the meantime.

Waste Reduction in a Board Shop

Waste in a board shop usually appears as waiting, movement, over processing, rework and defects, and very little of it looks like the scrap in a skip. Panels waiting between steps, operators walking to find tools, extra inspection that exists because an upstream process is unreliable, and rework that is so routine it is scheduled are all waste, and all of them cost time without adding anything the customer pays for.

The most reliable way to find waste is to follow one job through the flow and record everything that happens to it. The record usually shows that the actual processing time is a small proportion of the total, and that most of the elapsed time is queueing. From there the candidates for waste reduction become obvious, and the biggest ones are usually waiting and rework rather than a process step that is too slow.

Employee Involvement

Employee involvement is the part that is easiest to declare and hardest to build. It requires three conditions: the people doing the work must know what the measures are, they must be able to change something without a long approval, and their suggestions must receive an answer whether or not they are adopted. A suggestion scheme with no authority and no response produces one round of ideas and then silence.

The most effective mechanism is usually a short daily or weekly review at the area, standing up, with a small board that shows the current measures and the open actions. It takes ten minutes and it does three things at once: it keeps the measures visible, it gives people a place to raise problems, and it makes the open actions visible so that the ones that are not moving become obvious to everyone rather than to one manager.

Measuring Whether It Works

Improvement should be measured with the same figures the shop already uses: first pass yield, on time delivery, scrap cost, lead time and rework hours. Inventing a separate improvement metric produces a second set of numbers that nobody trusts and that can be made to look good independently of the shop. The point of measuring is to show whether the last change produced a result and to decide whether to keep it.

The measures should also be visible to the area that affects them, which usually means one or two figures per area rather than a company scorecard. An area that can see its own scrap rate and its own rework hours can connect a change to a result, and that connection is what makes improvement self sustaining. The quality data and the fabrication notes already contain most of what is needed.

Practical Rules

Start with a narrow area, change something within the week, write the change into the standard, and show the measure. Repeat in the same area until it is boring, then move to the next one, because the habit matters more than the selection of the first project.

Give the people who do the work the measure and the authority, and answer every suggestion. Most of the gain in continuous improvement comes from a large number of small changes made by the people closest to the work, supported by a process flow that shows what actually happens, and a shop that keeps that loop running for a year will outperform one that runs a single large project and then returns to business as usual.

FAQ

What is the difference between kaizen and a project? Kaizen is small, frequent and local, made by the people doing the work. A project is larger, rarer and usually owned by engineering or management.

Why does improvement not stick? Usually because the change was never written into the standard work, so the next person reverts to the previous method.

Which measure should be used? The ones the shop already trusts: first pass yield, on time delivery, scrap cost, lead time and rework hours.