Our company runs 5S. Most kilang in this industry do. Sort, set in order, shine, standardise, sustain — the audits, the red tags, the photographs on the board outside the office.
Kaizen sits inside that programme for us. Small improvements, made continuously, by the people doing the kerja. Not a project. Not a budget. Something everyone is expected to contribute to.
This is my contribution. Two small apps, built for the plant I work in, and what I have learned about where they actually fit.
Seiton, for something you cannot pick up
The second S is the one I keep coming back to. Set in order. A place for everything and everything in its place, arranged so anybody can find it without having to ask.
We are good at this for physical things. Shadow boards in the bengkel. Labelled bins. Spares sitting where the shelf says they are.
We were not good at it at all for what we knew. Ask what went wrong with a particular press last year and the answer sat in three places — a logbook nobody could read, a WhatsApp group four hundred messages deep, and the memory of a man who had since transferred out. Every item existed. None of it could be retrieved.
That is a Seiton failure. It just does not look like one, because nothing is lying on the floor.
What the apps actually are
MaintenX gives every machine its own record — what has gone wrong with this one, how often, and who fixed it. One place, searchable, on the phone the crew is already carrying.
Ullage Bunker is smaller still. Three sounding dips in, net tonnage out. It replaces a printed chart and a calculator balanced on top of a bunker.
Neither is clever. Both remove a step where the right answer used to depend on finding the right piece of paper.
Sort and shine come along for free
Sorting takes care of itself. The printed chart and the calculator are no longer needed on the bunker. The logbook is no longer the record of anything. Two fewer things to keep, and to keep track of.
Shine is the more interesting one. We already accept that cleaning is inspection — you find the weeping gland while wiping the machine down. What we never had was anywhere for the finding to go. It got mentioned to somebody, and then it evaporated. Now it lands in that machine's record and is still there in six months when the same thing goes again.
The fifth S is the one that fails
Everybody's 5S holds for the first three months. Then the audit gets softer, the board goes stale, sekali-sekala someone remembers to update it, and the standard quietly becomes whatever people can recall of it.
Sustain is not a motivation problem. It is a design problem. If following the standard is harder than not following it, it will not hold, and no amount of briefing changes that.
So I set myself one rule before anything else: under a minute, or the juruteknik will not use it — and an improvement nobody uses is not an improvement. Most of the ideas I liked did not survive that rule.
This is where convenience stops being a nice extra. Convenience is the reason the standard holds at all.
A few decisions came out of that. Tag a fault electrical and only certified chargemen appear in the list — the wrong assignment is never offered, so it cannot be made. Whoever did the work cannot close their own job; somebody else confirms it. A job can be assigned berdua, because that is how hot work is actually done here, and a system that cannot express it will simply be worked around.
A briefing about any of those decays in three months. Built into the form, they do not decay at all.
What it has been worth
I want to be careful here, because this is the part of a writeup that usually gets inflated.
A screw press that used to fail about every 38 days started failing every 20. Nobody reported that, and nobody was at fault for not reporting it — each work order looked ordinary on its own, and you would have to hold six months of them in your head at once to see the gap closing. The record showed it plainly. That is the clearest single thing I can point at.
What I cannot give you is a proper before and after. There is no baseline, because the old record was unreadable — which is the whole reason I built this in the first place. Anyone starting from where I started will have the same gap, and it is better to say so than to dress up a number.
What I can say plainly is that twenty people use it and not one of them asked for it. For 5S that is the measure that counts. Not whether the standard exists on paper, but whether it is being followed on a Sunday night when nobody is auditing.
Where this goes
I am not claiming a transformation. This is 5S applied to information, with kaizen's rule about who does the improving — the person doing the work, in small steps, at no cost beyond the time to build it.
What I did not expect is that it makes the next improvement easier. A maintenance interval can now be set from this machine's own history instead of a figure copied out of a manual. That was not available to us before, and I did not build it deliberately. It fell out of having the record.
Small improvements, continuously, by the people doing the work. That was always the instruction. These apps are just where I could make mine.