Four Months of Saying the Same Thing in the Same Meeting About a Staircase

 The escape staircase was built about ten years ago. Structure only, no services, sealed up and left waiting for whoever came back to finish the job.

We're the ones who came back. And the clear headroom under that existing roof slab is 2.1m, which was plenty when it was bare concrete, and is turning into a proper headache now that real cables have to run through it.

I've been raising this in CCSM meetings for months. It's still not resolved.

What CCSM Actually Is

Coordinated Combined Services Model. Every services-wide contractor builds their own model with real installation detail in it, then everything gets combined into one and reviewed together before anyone installs anything.

On some projects the main contractor runs this. On ours, the client's BIM consultancy role went to one of the SWCs, so they handle clash checks, model aggregation, SEM drawing generation, and coordination between the contractors.

I sit on the main contractor side. My job in that room is mostly saying uncomfortable things early. You can't add openings after our walls go up. There's no hacking once this is cast. Everything in this model was provided against the SEM drawings you signed off on. We've already built that section. That shop drawing is approved, so a change now means extra review and extra cost, and it won't be coming out of our scope.

Everyone's in the same room with a projector. Every SWC, the client, us. Fortnightly, sometimes monthly. Attendance is close to full.

The Deadline Everyone Misses at Least Once

There's a hard cut-off for model uploads before each round. Anything late doesn't make the clash check, and the notice goes out formally, upload late and whatever breaks afterwards sits with you.

The actual meeting is much softer than the notice. Someone missed it, apologises, says next round. Fine, we say, just fold in everything from today when you do.

That works because nothing's built yet. Everything's still movable.

Once construction starts, the same conversation goes somewhere else entirely. Openings are provided, structure's in, and someone asks for one more penetration. Now it's cost and programme, and the tone in the room changes fast. Everyone knows exactly where that line sits, which is why the polite version only lasts as long as it does.

Why the Staircase Still Isn't Fixed

Back to the headroom. When the SWCs first modelled their cable containment, it all looked fine. Nothing flagged.

Then they modelled the supports. Rods, brackets, hangers, the things that physically hold the containment up. Put architectural finishes on top of that and 2.1m simply isn't there anymore.

That gap between modelling the cable tray and modelling everything holding the cable tray has caused more late problems on this project than anything else. Everyone models the service. Far fewer model what it takes to actually install it.

And this staircase is existing structure, built a decade ago, sitting there fully cast. So the easy fix, hack the roof slab, carries risk I'm not remotely comfortable with.

The architectural contractor drew it up as though that hacking was already agreed. Just mark it and move on. I opened that drawing at my desk and said something out loud that I won't repeat here, because marking a hack on a ten-year-old structural slab isn't a drafting decision, and it definitely wasn't theirs to make.

I've raised it every meeting since. I'll keep raising it. There's a specific tiredness that comes with being the person who says the same thing at the same meeting for the fourth month running, and knowing you'll say it again next time.

The Site Doesn't Look at the Model

Here's the part nobody says loudly enough. Site installs from drawings. Not from the coordinated model.

So whoever gets into a space first installs to their own drawing. Whoever turns up second finds the room gone and improvises, drops the run lower, adds a U-shaped detour around whatever's already there. They're not being difficult. They physically can't put it where the model said, because someone fitted theirs tighter than the model expected.

Then that dropped run eats the headroom, or fouls a ventilation shaft opening, and we're right back in the same argument from a different direction.

That's why support spacing and rod positions matter so much in CCSM. Without them you're looking at a model of what somebody hopes will fit, and I've stopped calling those coordinated.

One Line, Many Contractors

The way this project is set up makes it unusual. The SWCs are appointed line-wide across the whole corridor, while main contractors differ by package. So the same services contractors are working with several different builders simultaneously.

An issue on someone else's package therefore gets highlighted to everyone. That's genuinely useful. A problem three packages away lands in front of us before we walk into it, usually with a solution attached.

The catch is that a fix that suits one contractor can be awful for another. If they haven't built that section and we have, or their solution costs us money it doesn't cost them, we end up arguing about a decision that was effectively made somewhere else by people who weren't thinking about our site.

The consultant situation has been messy too. The client's original consultant took it to final submission, got approval, then stopped. Someone else picked up the CCSM lead afterwards, and we've been cycling revisions ever since. Dash, A, B, and still going every fortnight.

The Chasing Problem

Attitude here is generally good, better than I expected coming in. Singapore's BIM maturity is high, and the client actually understands what they're looking at, which changes how seriously everyone treats it.

The friction comes from those line-wide appointments. If a site doesn't chase, that site slides down the priority list. Models arrive late. Occasionally someone insists they've updated when they haven't, or that it's uploaded when it isn't.

I don't argue about that out loud anymore. I check timestamps. When did you upload this, is this the latest, because what's sitting in the folder isn't. You might well have done it, but you haven't shared it, so nobody else can see it.

Said politely, that usually lands. When it doesn't, it goes into a formal email with the client, every SWC and the contractor copied. That works every time, which is a slightly depressing thing to have learned.

Nobody enjoys escalating something that should've been a two-minute conversation. I've stopped feeling bad about it though, mostly because by the time I send that email, the polite version has already been tried three times.

Compared to what I remember of Korea six years ago, before I left, this is a different world. Back then BIM was mandatory only above a certain project value, plenty of teams did it grudgingly, models drifted away from drawings, and issue reports came back with some version of site will sort it out. It's probably improved a lot since. That's a whole separate post I keep meaning to write properly.

Most of what I actually do in that room isn't even about protecting our structure from the SWCs. It's catching the thing two of them are about to do to each other, in a section nobody's opened yet. Some meetings I barely speak. Some meetings I'm the only one talking, and I walk back to my desk afterwards genuinely unsure whether any of it landed.

Related reading:

They Signed Off on the Opening, Then Came Back Wanting It Bigger

Squinting at a Decade-Old Scanned Drawing Until I Found the 100 Millimeters Nobody Caught