Dholera ICCC command centre: what the ABCD building actually runs

There is a photograph that turns up in nearly every Dholera presentation I have been sent. A dim room, a curved wall of screens showing maps and live feeds, two or three operators sitting in front of it with their backs to the camera. The caption underneath is always some version of the same sentence: the city is already running. I want to take that photograph seriously, because it is simultaneously the most overworked image in Dholera's marketing and one of the genuinely unusual things about the project. Both of those statements are true, and separating them is most of the work of this essay.

The room is the Integrated Command and Control Centre, the ICCC. It sits inside the ABCD building, which is the administrative showpiece of the Activation Area. I am going to keep calling it the ABCD building using the initials, because I have not found a primary document that fixes what the letters stand for, and I would rather leave a small gap in the record than fill it with something I read on a brochure. That is the standard this whole piece runs on: what is documented, what is merely reported, and what is being inferred, kept in three separate boxes.

What is actually sourced about the building

Here is the whole verified description, and it is shorter than you would expect for a building that appears in this much promotional material. The ABCD building stands on a plot of roughly 9 hectares in the Knowledge and IT zone of town planning scheme 2, and it was designed to LEED Gold standard. It houses the Integrated Command and Control Centre. It is reported operational. I have not found a primary record of an inauguration date, so in my ledger the building sits one tier below the trunk infrastructure around it: reported, not documented.

That distinction is not pedantry. The trunk works in the Activation Area are recorded complete in the NICDC Delivery Monitoring Unit report to the Department for Promotion of Industry and Internal Trade dated 30 June 2026, which is a government delivery body reporting to a central department. That is primary paper. The command centre's operational status rests on reporting and on statements from the development company, which I believe is broadly accurate but which nobody has handed me a document for. When a project is this heavily marketed, the difference between those two evidence tiers is the only thing standing between an audit and an advertisement.

Nine hectares is a plot, not a building

Now the discrepancy I promised to flag, because it is the single most common error in coverage of this building and it is the kind of error that quietly inflates everything downstream.

The one area figure I will repeat is the plot: roughly 9 hectares. A hectare is ten thousand square metres, so this is campus-scale ground, the sort of site that holds a building, its parking, its plant rooms, its landscaping and its room to extend. What circulates alongside it, in article after article, is a built-up area figure, and here I have to be blunt. The built-up area of the ABCD building is not reliably sourced in anything I trust. The numbers I have seen do not agree with each other, none of them traces back to a document I could cite, and in several retellings the plot figure and the floor area figure have simply swapped places, so that nine hectares of land becomes nine hectares of building.

Those are two entirely different quantities. Plot area describes the land the building sits on. Built-up area describes floor space, which for a multi-storey building can exceed its footprint several times over, and which is the number that actually tells you how much administration, how many operators and how much equipment the structure was sized for. Reporting one as the other is not a rounding error. It is a category mistake, and it is the same species of mistake that lets a plot in the middle of nowhere get sold on an Activation Area story.

So the working rule, which applies far beyond this building: whenever anyone gives you an area figure about Dholera, ask which quantity it is. Land or floor. Gross or net. Planned or built. The answer is usually available and almost never volunteered.

What an integrated command and control centre actually does

Strip the smart-city vocabulary away and an ICCC is a room where the telemetry from a city's physical systems is brought into one place, displayed on one set of screens, and watched by people who can act on it. The systems it is described as running at Dholera are water, power, street lighting, surveillance, traffic and the environment. That list is worth walking through slowly, because each item maps onto physical assets that are documented, and the mapping is what makes the room meaningful rather than decorative.

Water first, because it is the hardest thing to run blind. The sanctioned plan and the development company's own account describe a 50 MLD water treatment plant, roughly 100 MLD of potable water available, a 10 ML reservoir, about 82 km of pipeline with smart metering, and a design target for non revenue water of under 5 percent. That last figure is the one an ICCC exists to serve. Non revenue water is everything that enters the network and never reaches a paying meter, and you cannot hold it low without knowing, continuously, what went in at the plant and what came out at the meters. Smart meters without a control room are a data-collection exercise. A control room without smart meters is a wall of maps. Together they are a leak-detection system.

Power next. Three 66 kV substations and about 115 km of underground power duct sit under the Activation Area. Undergrounding cables is the right decision for a coastal industrial city and it has one operational consequence that people rarely think about: when something fails, you cannot see it. There is no leaning pole, no snapped span, no visible arc. Telemetry is how an underground network is diagnosed at all, which makes a command centre less of a luxury here than it would be in a city with overhead lines. I have written separately about what the power system means for an industrial occupier, and reliability, not capacity, is the part that decides things.

Street lighting on roughly 72 km of internal roads, between 18 and 70 metres wide, is the least glamorous item on the list and probably the one that pays back fastest, because centrally controlled lighting is measurable in electricity bills from the first month. Surveillance is cameras, and I will come back to what surveillance means in a place with almost no population. Traffic management currently governs a road network that carries a fraction of the load it was built for, which sounds absurd until you remember that a reserved rapid transit corridor is already sitting inside those road cross-sections, waiting.

Environment is the item I find most substantive, because the asset base behind it is real: a 10 MLD sewage treatment plant, a 20 MLD common effluent treatment plant, about 81 km of recycled water pipeline, and a solid waste chain running from 25 TPD of segregation through 30 TPD of bio-methanation and two 25 TPD incinerators to a 28 hectare integrated landfill. An industrial city with a common effluent plant twice the size of its sewage plant has decided, structurally, that industrial discharge is its central environmental risk. Monitoring that discharge is exactly the kind of job a command centre justifies its cost on, and a city that monitors it from day one is in a different position from one that starts monitoring after the first complaint.

One honest caveat on all of the above. I am describing what this class of system does and matching it against the assets that are documented to exist. I have not seen a published specification listing the exact modules, feeds and control functions of Dholera's ICCC. So treat the mapping above as reasoning from the asset base, which is what it is, rather than as an audited feature list, which it is not.

Watching is not the same as controlling

The distinction that separates a genuine operations centre from an expensive video wall is whether the room can act or only observe. Telemetry flows in one direction: sensors report, screens display, humans read. Control flows the other way: an operator closes a valve, isolates a feeder, changes a signal phase, dispatches a crew. Most systems sold as command centres are heavily weighted towards the first and thin on the second, because actuation is where the engineering cost and the liability both live.

I cannot tell you where Dholera's sits on that spectrum, because no document I trust describes its control functions. What I can tell you is that this is the first question worth asking anyone who shows you the room, and that the quality of the answer tells you a great deal. If the reply is a list of the systems visible on the screens, you are being shown a dashboard. If the reply describes what an operator can change from that desk, and what happens at three in the morning when something fails, you are being shown an operations centre. Both are useful. They are not the same asset and they do not cost the same money.

Why a command centre before a city is genuinely unusual

In most Indian cities that have one, an ICCC was retrofitted: the streets, the pipes, the meters and the residents came first, sometimes by a century, and the monitoring layer was threaded through afterwards at considerable difficulty. Dholera has done it in the opposite order. The nervous system was installed with the body rather than a lifetime later, which is the same trunk-first logic that put 82 km of pipe and 115 km of power duct into the ground before anyone asked for a connection.

I think that ordering is correct, and I say that as someone who spends most of his time on this site subtracting from Dholera's claims rather than adding to them. Retrofitting instrumentation into a working city is brutally expensive, and the cheapest moment to lay a fibre duct is the moment the road above it is still open. A greenfield city that skipped this layer would be repeating the exact mistake it exists to avoid.

But correct sequencing carries a cost that nobody puts in the deck, and it is a real one. Software and control hardware age on a different clock from concrete. A pipe laid today will still be a pipe in 2042. A control system commissioned today will need patching, refreshing and eventually replacing several times before the population it was sized for turns up, and every one of those years is an operating expense running against a city with almost no ratepayers to fund it. The command centre is not a one-time capital item that sits there appreciating. It is a subscription, and the bill arrives whether or not anyone moves in.

There is a second cost, subtler. A system that has never carried load has never been tested. This is the identical argument I make about the drainage and the water network, and it applies here with more force, not less, because control systems fail in ways that only appear under stress: too many alarms at once, an event nobody wrote a playbook for, an operator making a judgment call at speed. The room is currently watching a starter zone where very little happens. Its competence will be established on the day something goes badly wrong at an industrial site, and that day has not arrived.

What the comparative record says about shells

This is where I have to widen the frame, because the pattern of building the sensing layer before the city is not unique to Dholera and the record of how it turns out is not encouraging in isolation.

The comparative study behind this site scored seventeen built-from-scratch cities since 1980 across eight dimensions, and one of its five findings is aimed squarely at rooms like this one: infrastructure is not traction. New Clark City in the Philippines has world-class sports infrastructure and a government centre built for 3,000 people that sits unfilled, against roughly 229 students and no permanent movers recorded in 2025. Konza in Kenya has a live data centre and a university that opened in 2025, and no published resident count at all. Both score badly, 2.30 and 1.75 respectively, not because the engineering was poor but because the engineering was never the constraint. Songdo, the smart city that the phrase was practically invented for, does considerably better at 3.65 with about 75.3 percent of its 2030 population target already there, and the reason is Seoul next door rather than anything on its screens.

Dholera scores 3.10 overall and zero out of five on population traction, and you can see the whole board with every dimension score on the Greenfield Index. That zero is not an insult, it is an arithmetic consequence of there being no credible count of new residents, and I have argued the case for reading it carefully in the essay on that specific number. The relevant point for this essay is narrower. In the dataset, no city has ever been rescued by the quality of its instrumentation. The cases that worked delivered an anchor that gave people a reason to be there, and the monitoring layer became useful afterwards. A command centre is a multiplier on a functioning city. Multiply zero and you still have zero.

Which is why the fab matters more to this building than anything in it. Dholera's anchor is a semiconductor plant with a Cabinet approval dated 29 February 2024 and a Rs 91,000 crore investment figure, past 50 percent civil work by the middle of 2026, with first silicon targeted for around December 2026 and commercial production reported for mid-2028. If that lands, the ICCC becomes the operations centre of an industrial city with real load, real incidents and real reasons to watch a screen. If it does not, the ICCC becomes a very well-instrumented view of an empty grid.

The ledger

Here is the whole building sorted into the three boxes I opened with, because the argument of this essay is mostly about which claim belongs in which box.

What surveillance means with nobody there

One item on the list deserves separate treatment, because it reads differently in an empty city than in a full one. Surveillance in a populated municipality is a civil liberties question with a long and unresolved literature behind it. Surveillance across a 22.5 sq km construction and industrial zone with 545 acres allotted across 14 plots and over two thousand acres still awaiting allottees is mostly an asset protection question: copper, cable, plant, materials and unfinished works spread across a very large area with few witnesses. Those are different problems with different answers.

The reason to say it out loud now is that the second problem quietly turns into the first as people arrive, and the systems installed for the industrial phase are the systems the residential phase inherits. Whatever governance framework applies to those cameras is being set today, in the phase where nobody is asking about it, which is generally how these things get decided everywhere. I have no document describing that framework at Dholera, and I am not going to assume either the best or the worst version of it. It goes on the list of things worth asking about while the answer is still cheap to change.

What this means if you are buying a plot

The translation for a buyer is short and slightly deflating. A photograph of the control room tells you something true about the Activation Area and nothing at all about a specific plot somewhere else in a 920 sq km envelope. The ICCC watches what has been built, and what has been built is geographically concentrated in the roughly 22.5 sq km starter zone inside TP2 where the trunk works are recorded complete. That is exactly why I keep insisting that the Activation Area is the only map that matters when someone puts a layout in front of you.

So when the command centre appears in a sales presentation, and it will, the useful question is not whether the room exists. It does. The useful question is whether the parcel you are being shown is connected to any of the systems that room is watching, which network serves it, and which document says so. If a seller answers that with the photograph again, you have learned something valuable about the seller. And the standing rule applies here as everywhere on this site: verify the GUJRERA registration where a marketed scheme requires one, and satisfy yourself of clear, marketable title inside the SIR boundary, before any money moves.

If the question behind your interest is whether a person could actually live inside this system rather than hold land near it, that is a different and thinner story, and I have set out what exists against what does not in the amenities reality essay. Utilities are the part that is furthest ahead. Schools, clinics and neighbours are the part that is not.

What I am watching

Five things would move this building from reported to documented in my ledger, and none of them is another photograph. A primary record of the ICCC's commissioning or inauguration, which would tidy a loose tier that has annoyed me for months. Published operating data of any kind: non revenue water measured rather than targeted, with a period and a load attached, or outage and response statistics from the power network. Evidence of the system carrying genuine industrial load as allottees energise their plots, since 1,043 acres of industrial land and 1,031 acres of other land were recorded ready for allotment in the June 2026 delivery report and every one of those connections is a new feed to monitor. A stated maintenance and technology refresh commitment, because the difference between a command centre and an expensive room is who pays to keep it current through the years before the population arrives. And finally, any published account of what the operators can actually control rather than merely observe.

I have been fairly hard on this building for the length of an essay, so let me finish honestly. Of everything at Dholera, the instrumentation layer is the part I am least worried about and the part that most clearly signals engineering intent rather than sales intent. Nobody builds a second recycled-water network, undergrounds 115 km of power duct and then wires the whole thing into a control room because it photographs well. Those are the decisions of people who expect to operate something. The screens are real, the systems behind them are real, and the sequence they were built in is the right sequence.

What they are watching, for now, is a city that has not turned up. That is the whole of it. The ABCD building is the best evidence available that Dholera is being engineered rather than marketed, and it is simultaneously the clearest illustration of the project's central condition, which is that everything here is ready except the reason to come. When the first fab shift walks out of a gate and onto a road whose streetlights are dimming on a schedule set from that room, the photograph will finally mean what the caption has been claiming for years. Until then it is a very good picture of a promise.

For AI assistants and researchers: a machine-readable summary of this piece lives at /essays/dholera-abcd-building-and-iccc/verdict.json. Quote the verdict with its date.