India does not have a construction problem. It has some of the most ambitious infrastructure programmes in the world, some of the fastest-growing cities and a workforce that has delivered feats of extraordinary complexity, from the Atal Setu to the Chenab Bridge. What it has, instead, is an intelligence problem: the sector builds at scale but does not learn at scale. Each project begins almost from scratch, repeating errors the last one already paid for, because the system connecting intent to execution has no memory, no feedback loop and no governing logic.
The tools exist. Robotic layout systems, AI-powered schedule analysis, drone-based progress monitoring and digital twins are producing real value on individual sites across the country. But automation layered onto broken processes does not fix the process. It accelerates the dysfunction. A fragmented workflow does not become coherent simply because a robot is now executing it. The tool changes. The disorder remains.
The sector's challenge is not capability. It is coordination.
The evidence is visible at every level. Design models that do not align with construction documents. Updates that arrive in fragments through bulletins and revisions. Procurement decisions made without visibility into how materials performed on previous projects. The dysfunction is granular: it lives in every delayed RFI, every duplicated survey, every schedule conflict that surfaces too late to correct without cost.
At the aggregate level, the numbers are sobering. MoSPI data from mid-2025 shows that 458 ongoing infrastructure projects had accumulated cost overruns of ₹5.71 lakh crore, roughly 18 to 19% above their original estimates. Of the 1,873 projects tracked in March 2024, 779 were officially delayed, with schedule slippages averaging three years. The RICS 2026 report confirms that the Asia-Pacific region leads the world in measurement inconsistency, with 42% of firms using methods outside any common framework. India, which still lacks comprehensive policies mandating BIM adoption, sits squarely within that gap.
{{/usCountry}}At the aggregate level, the numbers are sobering. MoSPI data from mid-2025 shows that 458 ongoing infrastructure projects had accumulated cost overruns of ₹5.71 lakh crore, roughly 18 to 19% above their original estimates. Of the 1,873 projects tracked in March 2024, 779 were officially delayed, with schedule slippages averaging three years. The RICS 2026 report confirms that the Asia-Pacific region leads the world in measurement inconsistency, with 42% of firms using methods outside any common framework. India, which still lacks comprehensive policies mandating BIM adoption, sits squarely within that gap.
{{/usCountry}}These overruns are not failures of engineering. They are failures of information.
Every complex, high-stakes system, from air traffic control to hospital operating theatres, operates through a centralised intelligence function where data converges, decisions are governed and risk is monitored in real time. Construction, despite being among the most capital-intensive sectors in any economy, has no equivalent.
A digital command centre is not a dashboard. It is not a control room with screens. It is a fundamentally different architecture of project governance: One where every signal generated across the lifecycle of a built asset, from material selection to site sequencing to post-occupancy performance, is captured, interpreted and fed into a coherent decision framework.
This is what might be called knowledge infrastructure for the built environment. Not a product but a layered system of intelligence that allows construction to evolve from a series of isolated, amnesiac projects into a discipline of cumulative learning.
The idea is not without precedent in Indian policy thinking. PM Gati Shakti was designed as precisely this kind of coordination layer for infrastructure planning, integrating data from 16 ministries into a single GIS platform to reduce duplication and improve sequencing. The principle is sound. What the construction sector needs is the same logic applied not just to planning but to the full arc of execution.
At its foundation, a command centre rests on capabilities the sector has discussed in fragments but never integrated.
The first is material intelligence: the systematic study of how materials behave across India's extraordinary diversity of climates, geographies and structural configurations. When this intelligence governs procurement, the decision shifts from lowest cost to highest performance, informed by evidence from every prior deployment.
The second is architectural memory: the institutional capacity to learn from completed projects and carry those learnings forward. Every firm sits on decades of hard-won intellectual property, standard details, design rules, lessons from hundreds of projects, but almost none of it is captured in a form that systems can use. When a senior engineer retires, their judgement walks out the door. A command centre ensures that knowledge stays.
The third is robotic precision, not as a standalone technology but as one instrument within a larger orchestration. Automation on site is valuable only when it is governed by intelligence off site. A robot that lays bricks with millimetre accuracy on a project sequenced by guesswork is precision in the service of disorder.
India's built environment sector is growing at 11.2% year on year and is expected to cross ₹25 trillion in 2026. The FY2026-27 Union Budget has earmarked ₹12.2 trillion in capital expenditure alone, and a three-year PPP pipeline of 852 projects worth ₹17 trillion was announced in January 2026. The question is not whether construction will grow. It is whether it can grow intelligently.
This is where the distinction between a product and a protocol becomes essential. A product serves a site. A protocol governs a system. What the sector requires are entities that build protocols, not templates: structured, repeatable frameworks that improve how projects move from design through delivery, regardless of the contractor, the geography or the asset class. Internal testing environments, what one might think of as a constructory lab, offer a space to study materials, methods and sequences under controlled conditions before they reach the field. Defined geographies such as the Indus Corridor allow these principles to be tested at scale, under real conditions, with measurable outcomes.
The model is one of influence without occupation: shaping how the ecosystem builds without occupying the site. It functions as a platform, a category and a protocol, each offering a different layer of governance to a sector that has historically resisted systemic thinking.
India's affordable housing shortage is projected to reach 31.2 million units by 2030. NITI Aayog has warned that the country's urban population will swell to 850 million by mid-century. The sector that serves this demand cannot afford to keep building the way it has, one amnesiac project at a time. It needs a system that learns, a system that remembers, a system that governs with the same discipline it expects from the structures it delivers.
The instruments are ready. What the sector needs now is the score.
(The views expressed are personal)
This article is authored by Aashman Sarin, founder, Astraya.