What construction has never digitalised — and the PLOT method to fix it.
By Mikhail Khrushchev · First published 8.11.2026, bimsoft.qa
I have spent my career between engineering and construction. I adopted BIM early, learned it deeply, and built software that made engineering work better.
I spent that career on sites too, in several countries — and there the picture was different. Engineering is digital, precise, advanced. Fifty metres away, the site still runs on paper and on fragmented registers: each kept by a different department, in a different subcontractor’s office, in a different format. They get compared after the collision, over the phone.
Everything in those registers is the same kind of thing — temporary work. Together it forms a whole layer of the project that nobody models, nobody shares and nobody keeps: a hidden temporary layer.
The engineering model shows the goal. Nobody shows the site.
Let me explain what I mean.
1. An empty field and a perfect model
A big industrial project has one complete picture of its final goal: the engineering model. Everything that was designed, calculated and approved is in it. You can turn it, cut it, measure it. It is the ultimate description of what thousands of people are working to create.
Then you drive to the site. And on a new project, you see an empty field.
Between this field and that model there are several years. Hundreds or thousands of people. Millions of man-hours, and an enormous amount of work that the model says almost nothing about:
- excavations
- cranes and lifting operations
- laydown and storage areas
- scaffolding
- camps, offices, temporary shelters
- welding, testing, energisation — dangerous operations of every kind
Every one of these things is real. Every one takes space. And every one is temporary. A laydown area lives three months. A scaffold lives six weeks. A lifting zone lives one afternoon. They appear at whatever spot was free at the moment, they get in each other’s way, and then they disappear.
So every construction site has two layers:
The permanent layer — what we are building. It is modelled, checked, approved, versioned. An entire industry called BIM exists to manage it.
The temporary layer — everything the people do to build it. It is managed by memory, radio, and Excel.
We digitalised the goal. We barely digitalised the work.
2. The temporary layer is silent
Think about how a site actually runs today.
A scaffold crew starts erecting on Monday. Where is that recorded? Maybe in a request form. In which system? Whose? Can the crew working twenty metres away see it? Can the crane coordinator see it? Will anyone see it in two years, when the next project asks how long scaffolding really takes?
Now the same question for the laydown area that grew into the access road. The radiography shot that needs the area cleared tonight. The excavation across the only walking route. The trailer somebody parked in a convenient spot.
The honest answer is: it is recorded — somewhere. A large site is not undocumented. There is a document control system. There are permit systems. There are registers — many of them: one per discipline, one per division, one per subcontractor. The scaffold request is probably in one of them.
But look at what those records can actually do. Each register sees only its own corner; none of them sees the others, so nobody can put the claims side by side. Nothing checks one claim against another, so a false or outdated line sits in a register as comfortably as a true one — contested by no one, until it meets reality. And the records themselves are, in the best case, scans locked inside PDFs. They are not useless: after a major incident, investigators dig real details out of them, and now and then an attentive reviewer catches a clash before it happens. But that is their ceiling — one document, one careful reader, one case at a time. As a source of insight they give nothing: you cannot ask a question of a thousand scans.
So the temporary layer is registered — and still silent. A record that cannot be combined, queried, or contested is a record that cannot speak. Everyone on site works inside their own small circle of visibility — a fog of war, in which whole crews appear out of nowhere twenty metres away.
And the morning meeting — the most important hour on any site — is people reconstructing this picture from memory, every day, out loud, and arguing about whose memory is right.
3. “Just use drones” does not work
The common answer to this is imagery: drones, satellites, 360° cameras. I like these tools. But they have two limits that no camera will ever fix.
They cannot see inside. Work under a roof, in a trench, inside a building — invisible. That is where the difficult work usually is.
They only show the past. A photo tells you what the site looked like when it was taken. It says nothing about next Tuesday.
And next Tuesday is the whole point. A photo answers “how does the site look?”. The site needs to answer “who will be where, doing what, next to whom — and where does that collide?”
You cannot photograph the future. You can only register it.
4. The temporary layer kills
If this were only about productivity, it would still matter. It is not only about productivity.
On 23 March 2005, at the BP refinery in Texas City, an explosion killed fifteen people and injured more than 170. Most of those who died were not process operators. They were contractors in temporary trailers, placed near a process unit — in a convenient spot, for a short time.
Those trailers were in no engineering model. No work package. No schedule. Somebody decided where to put them, and that decision was invisible to anyone who could have judged the risk.
The industry’s answer was a standard — API RP 753 — that says where you temporarily place things is a safety-critical decision.
Most serious site incidents follow this same shape: two activities meet in the same space, and neither knew about the other. A lift above a working crew. Hot work near a store. An excavation under a walking route. The safety world calls it SIMOPS — simultaneous operations. I call it what it is: the temporary layer colliding with itself, in the dark.
5. Other industries already solved this
Here is the strange part. This problem is not hard. Other industries solved it long ago, and their solutions all look the same.
Aviation has the NOTAM: a temporary hazard or restriction, with a place, a time window and an owner, published in advance to everyone it might affect. It has run worldwide for decades.
Railways book possessions: a section of track, a time window, an owner — agreed centrally, conflicts resolved before the night, not during it.
Utilities register street works: location and dates, permitted, so two companies do not dig the same street in the same month.
Even offshore oil and gas — construction’s own cousin — now runs live spatial systems that show every vessel and operation on one map and warn about conflicts before they happen. Supermajors pay for this today.
Look at the shape of every one of these solutions. A NOTAM, a possession, a street-works registration — each is the same object: a planned location, over time. These industries have been registering that object for decades. They just never needed one word for it.
Every one of them decided the same thing: a shared, forward-looking picture of who will be where is a safety requirement. Not a nice-to-have. A requirement.
Onshore construction — with more people, in denser space, doing more varied work than any of them — is the outlier. We still run our temporary layer via Excel registers and notes on paper scans.
6. What I believe: the PLOT method
So here is my position, as plainly as I can put it.
1. Every temporary thing on a site can be described the same simple way: a place, a date range, an owner. A scaffold, a permit area, a crane radius, a laydown, a hazard zone, an inspection — all the same shape. Where, when, who. I call this object a plot: a Planned Location Over Time. This one simple shape is why the work of thousands of people can live in one environment without chaos.
The word is not an accident. A plot is a piece of land — the most construction-native word there is. To plot is to plan ahead. You plot work on a chart — which is exactly what the register looks like. And a story has a plot — the site’s own narrative, recorded as it happens. Four meanings, and all four are the job.
2. A plot is registered before the work, not reported after. A record of the past is a report. A record of the near future is a planning tool. Only one of them prevents anything.
3. Plots are registered by the crews themselves — and this is the hard part. Offshore, every vessel broadcasts its position automatically. On a construction site, nothing broadcasts. A scaffold has no GPS. So the data can only come from people — which means registering a plot must be useful to the person doing it. It must be how a foreman claims his area, gets his permit, tells everyone he is there — and sees everyone who is already there. Nobody fills in data for management. People plot their works because it works for them, or they do not plot at all.
4. Plots belong in the engineering model. The permanent layer is already there, in 3D, with correct coordinates. The temporary layer must live on the same picture — so that the claimed area, the crane radius and the live system boundary sit exactly where they will physically be.
5. This is not another 4D simulation. The industry already has beautiful 4D tools, where a planner builds a simulation in an office — and it starts going out of date the same day. The difference I am arguing for is crew-authored, not planner-authored: a live record built continuously by the people who will physically stand in those spaces. Not a film about the plan. A register of reality, a few days early.
Put together: the temporary layer of a construction site, registered in advance by its own crews, as plots — located, dated, owned objects — in the engineering model. That way of working is what I call the PLOT method. And the result, combined with the permanent model, is a construction site digital twin. The permanent layer is the noun. The temporary layer is the verb.
7. What becomes possible
When the site’s works are plotted, things change on the ground.
Most conflicts dissolve at the moment of plotting. The foreman plotting a slab for Tuesday sees, right there, the crew that plotted it on Monday — and takes Wednesday, or the next bay, or makes one phone call. The collision never happens; it is negotiated away by the two people it concerned, days early. Only what foremen cannot settle between themselves goes to the weekly meeting — a short list of real disagreements, not a reconstruction from memory.
The model can show the future. Pick a date, and see the site as it will be that day — including things no drone can photograph, because they are not built yet. Your site, plotted.
Hazard zones become visible objects on the picture everyone shares, instead of lines in a file that exists mainly for the auditor.
The overlap of construction and commissioning becomes manageable. The most dangerous months of any project are when crews still build while systems come alive around them. “Who owns this space today?” finally has an answer you can see.
And — the quiet, biggest one — the site starts producing history. Which operations happened, where, in what order, how long they really took, what really blocked whom. Today this knowledge evaporates; it retires with the people. Plotted, it becomes data — the plot of the project, in the oldest sense of the word: the true story of how it was built. For the first time we could answer, with evidence, the question this industry has never been able to ask: how does a construction site actually work?
8. What I am not claiming
I want to be careful here, because this field is full of inflated claims.
I did not invent spatial conflict detection. Researchers formalised time-space clashes twenty years ago. Offshore operators run spatial coordination today. Digital permit systems with zone checks exist. 4D logistics tools model cranes and laydowns.
What does not exist is the combination: the whole temporary layer of an onshore site — including the work that belongs to no work package at all — registered prospectively by the crews themselves, in the 3D engineering model, in one environment that later hands the client not just an as-built model but the operational history of how it was built.
And I will not quote you a savings percentage. Nobody has measured this yet — which means anyone quoting a number for it is guessing. What I will do instead is measure it: openly, on real sites, with the metrics defined before we start and the results published whatever they say. Including failure, if that is the answer. The first honest numbers in this field are worth more than any claim I could invent.
9. An invitation
I am building this — the PLOT method and the software behind it — and I am doing it in the open.
If you run a site and this describes your morning meeting, I want to talk to you.
If you work in AWP, lean construction or HSE and you think I am wrong somewhere, I especially want to talk to you — the method will be better for it.
And if you want to be the first project with measured numbers on what the temporary layer actually costs, that is the most interesting conversation of all.
The model shows the goal. The temporary layer is how we get there. It is time to plot it.
Mikhail Khrushchev builds BIM software and site-coordination tools, and writes about the temporary layer and the PLOT method at bimsoft.qa. Contact: [email protected] / LinkedIn