Tasnim Kabir: Bangladesh excels at celebrating construction. New bridges, roads, power plants, treatment facilities, terminals and buildings generate photographs, speeches, ribbon-cuttings and political credit. The quieter questions—who will maintain these assets, how they will be monitored, what it will cost to keep them reliable, and what happens when they begin to fail—are usually postponed. Those questions can no longer be afterthoughts.

The engineering Bangladesh needs most in its next development phase is not simply more construction. It is better maintenance. This is not an argument against building. The country still requires stronger power networks, better roads, safer water systems, modern drainage, resilient ports, reliable hospitals and digital infrastructure. Building without maintenance, however, creates a costly illusion of progress. Assets can expand rapidly yet lose economic value if they deteriorate faster than they are managed.

The real test of infrastructure is not inauguration. It is performance over time. A road succeeds only if it remains safe and cost-effective under heat, rain, axle loads and flooding. A power system succeeds when electricity reaches homes, factories, hospitals and data centers reliably. A water treatment plant succeeds when pumps, sensors, dosing systems and pipes continue working under real conditions. A bridge succeeds when expansion joints, bearings, drainage and corrosion protection are maintained for decades. Maintenance is where infrastructure promises either survive or fail.

Most failures do not begin as sudden disasters. From experience in power generation, water infrastructure, industrial automation, reliability engineering and data-centre operations, they start as small signs: vibration, heat, leakage, pressure loss, repeated alarms, corrosion, and abnormal current, delayed inspections, missing spare parts, incomplete records or an unspoken maintenance backlog. These signals appear long before the public sees the breakdown.

Many institutions still operate reactively. Something breaks, and then repairs begin. A pump fails and emergency procurement starts. A road collapses and inspection becomes urgent. A transformer burns and replacement becomes a crisis. Reactive maintenance is expensive, disruptive and often unsafe. It is not a strategy; it is a symptom of weak asset management.

Bangladesh must shift toward preventive and predictive maintenance. Preventive maintenance means scheduled inspection and servicing before failure. Predictive maintenance uses data—temperature sensors, vibration monitoring, oil testing, electrical testing, SCADA trends, thermal imaging, pressure and flow records, maintenance logs and digital asset registers—to detect early warnings and estimate remaining life. These tools are standard practice in serious infrastructure systems, not futuristic luxuries.

The following reforms can be taken to overcome the issues.

Culture is main issue. Maintenance must be treated as engineering, not housekeeping. Too often it is seen as less prestigious than new design or construction. A maintenance engineer, who prevents a pump-station failure, avoids a transformer outage or extends a bridge’s life can save more public money than a new project announcement. The country needs to value the engineers, technicians, operators and field crews who keep systems alive after the cameras leave.

Finance is the vital factor. Budgets favour visible capital expenditure. Operation and maintenance budgets are less glamorous yet decide whether assets deliver value. Bangladesh should measure annual maintenance spending as a percentage of asset value across roads, bridges, power distribution, water utilities, drainage, public buildings, ports and hospitals. If those budgets are too low or released too late, deterioration is inevitable.

Data is the asset. Reliable asset registers are required for critical infrastructure. Every major utility and public agency should know what it owns, where assets are located, when they were installed, their condition, maintenance history and failure risk. Without this information decisions become political, anecdotal or crisis-driven. With it, agencies can priorities the assets whose failure would cause the greatest public harm.

Digital monitoring should be put in place. SCADA systems, sensors, GIS mapping, computerized maintenance management systems and digital twins move agencies from guessing to knowing. A digital twin is a virtual model that helps operators understand condition, simulate stress and plan interventions. Bangladesh need not digitise everything at once. It should begin with critical assets: substations, water plants, wastewater pump stations, major bridges, drainage pumps, hospitals, ports, economic zones and data centres.

Accountability must be ensured. Public agencies should publish simple performance indicators. For roads: pavement condition, bridge inspection status and repair response time. For power: outage frequency and duration, transformer failure rates and restoration time. For water: non-revenue water, pipe bursts, pumps failures, pressure and service continuity. For hospitals and digital systems: uptime and emergency-backup readiness. Existing data should be verified and made more publicly available.

Workforce must be kept in mind. Modern infrastructure cannot be maintained with outdated training. Bangladesh needs more technicians and engineers skilled in reliability engineering, electrical safety, automation, instrumentation, pump systems, rotating equipment, power distribution, water and wastewater operations, cyber security and data-driven maintenance. Universities and polytechnics should teach maintenance as a high-value discipline. Economic zones, utilities and public agencies should create structured career pathways for maintenance professionals.

Last but not least is procurement. Buying equipment is not enough. Agencies must evaluate life-cycle cost, spare-parts availability, local service capability, energy efficiency, and maintainability and documentation quality. A cheaper asset that fails frequently is not cheap. A project without a funded maintenance plan is incomplete. Every major infrastructure project should include a maintenance strategy before approval.

Bangladesh’s ambitions in manufacturing, digital services, logistics, energy and technology depend on infrastructure that works consistently. Investors look not only at what a country builds but at whether systems remain reliable years later. Better maintenance is also a public-interest issue. Citizens suffer when roads deteriorate, pumps fail, drainage collapses, hospitals lose power, digital services go offline or emergency repairs consume funds that could have been used more wisely. Maintenance is public protection as well as technical efficiency.

The country has entered a stage where the central question is no longer only “What will we build next?” It is “What we have already built, what condition is it in, and how will we keep it working?” A serious nation measures progress not by the number of projects it starts but by how long those projects serve the public safely, reliably and economically. Bangladesh has built with ambition. Now it must maintain with discipline. The next generation will inherit not the speeches made at inauguration but the roads, pipes, substations, pumps, bridges, data centres and public systems that either work—or fail—under the weight of daily life.

Writer: Tasnim Kabir is a Data Centre Engineering Operations Engineer (DCEO Engineer) at Amazon Web Services (AWS)