If I rely on a park operator for utilities, what happens to my GMP accountability and liability?

You cannot outsource GMP accountability and should not accept a model that pretends you can. The shared utility supply is distinct from your qualified systems: the park delivers power, water and effluent handling to an evidenced standard, while product-contact systems inside your envelope, such as your purified-water and Water-for-Injection loops, cleanrooms and process, remain yours and qualified by you. Demand the operator's qualification and monitoring documentation to feed your validation, plus service-level terms that make obligations and liabilities explicit.

Does sharing utilities between tenants create cross-contamination or regulatory risk?

Cross-contamination is a cardinal GMP failure, so the concern is legitimate. The shared layer is the non-product-contact commodity layer, such as bulk power, incoming water before final treatment and effluent after it leaves the process, while everything touching product stays dedicated, segregated and individually qualified per tenant. Tenants share the supply of a utility, not the utility where it meets their product. Each facility holds its own licences and is inspected on its own systems, so a shared campus does not merge compliance identities.

Is shared infrastructure actually cheaper than building your own utilities?

A shared model carries an operator return, so it is not automatically cheaper and should be modelled. But compare total cost of ownership. Self-provisioning means paying full capital cost for redundant capacity used only partly, plus its operating and management burden and the waste of idle redundancy, alone. Sharing spreads the redundancy capital and utilisation across tenants, so even net of margin the delivered unit cost can be lower, and it converts upfront capital into operating cost, changing the financing profile. The outcome depends on utilisation and operator efficiency.

What infrastructure does a pharmaceutical manufacturing plant in Nigeria actually need?

Q: Is shared pharmaceutical infrastructure a single point of failure for every tenant at once?

Pooling a dependency does concentrate it, so the answer turns on redundancy. Shared infrastructure can be more reliable, not less, because a park serving many tenants can afford redundancy no single manufacturer can justify alone: multiple generation sources, N+1 or N+2 configurations, duplicated critical systems whose cost is spread across the site. A single tenant almost always under-builds redundancy. The question to put to an operator is not whether it is shared, but to what redundancy standard and what guaranteed uptime.

Q: Isn't it an oversimplification to say GMP compliance is basically an energy load?

Yes, to a point. GMP compliance is overwhelmingly about quality systems, documentation, validated processes and people, and HVAC is a small part of the whole. The specific claim is narrower and about infrastructure: of the physical plant you must build and power to enable compliant manufacturing, environmental control is the dominant energy consumer, with cleanroom HVAC reported at roughly 36 to 67 percent of facility energy. The point is that the infrastructure layer compliance depends on is energy-dominated, which in a power-scarce market becomes the binding constraint.

A GMP pharmaceutical facility needs six infrastructure systems that most industrial plants do not: uninterrupted clean power, validated HVAC with classified air, purified water and Water for Injection, compliant effluent treatment, qualified cleanroom envelopes, and clean utilities such as clean steam, compressed air and nitrogen. In Nigeria, none of these can be reliably purchased as a service, which means a pharmaceutical manufacturer must build and operate a power utility, a water treatment utility and a waste treatment company as a precondition of making medicine. That, not the manufacturing equipment, is where most Nigerian pharmaceutical projects will fail post-feasibility study.
Author: Theodora Ogharanduku, CEO - Knectiv. Last reviewed: Aug 18, 2026

A key distinction ‐ built versus qualified

In general industry, infrastructure needs to work. In pharmaceutical manufacturing, it needs to work and be documented to have worked, reproducibly, under challenge.

Every critical system must pass design, installation, operational and performance qualification, and carry the documentation to survive a regulatory inspection. A water system that produces good water but cannot evidence it consistently is a non-compliant water system. This is why appointing a general industrial contractor to a pharmaceutical build is so expensive. The deficiency is not discovered during construction but during qualification, when correcting it is at its most costly.

Infrastructure that is built but not qualified has no value in this industry.

The six systems

1. Power ‐ continuity, quality, and why it dominates everything

Pharmaceutical power requirements are unusual not in quantity but in tolerance. An interruption is not lost production time, it’s a potential batch loss, an environmental monitoring excursion, a deviation investigation and a documented compliance event.

The requirement is therefore uninterrupted, stable, clean power with redundant generation, UPS protection on critical systems, and a supply architecture designed so no single failure can compromise a batch or a classified environment.

2. HVAC and classified air ‐ the largest energy load in the building

This is the system that makes pharmaceutical facilities unlike anything else on an industrial estate. Classified cleanroom environments require controlled particle counts, pressure cascades between zones, HEPA filtration, and continuously controlled temperature and humidity - maintained around the clock, frequently even when production is idle.

The energy consequence is decisive. Cleanroom HVAC has been reported in benchmarking studies to account for approximately 36% to 67% of total facility energy consumption, with HVAC the dominant contributor to energy intensity at pharmaceutical sites (ISPE). Pharmaceutical cleanrooms can consume up to 15 times more energy than commercial building systems, driven by the high air change rates required to maintain air quality (ISPE).

This is the causal link most Nigerian pharmaceutical business cases miss. GMP compliance is, in energy terms, largely an HVAC load. In a country where reliable power is the scarcest industrial input, the compliance standard and the infrastructure problem are not two risks, they are one risk. You cannot economise on air classification without economising on compliance, and you cannot economise on compliance without losing your market.

3. Water ‐ purified water and Water for Injection

Pharmaceutical manufacturing requires water treated to pharmacopoeial standard ‐ purified water for most applications, water for injection where sterile products are made. This means generation systems, storage and continuously circulating distribution loops designed to prevent microbial growth, with sanitisation regimes and ongoing monitoring.

Two site-specific factors drive cost - the volume required and the quality of the raw source. Poorer source water means more treatment capex and more operating cost, permanently.

4. Effluent treatment ‐ the disqualifying constraint

For API synthesis in particular, effluent is the constraint most likely to disqualify a site outright. Solvent‐bearing and chemically complex waste streams require treatment, and in some cases solvent recovery or incineration, to a standard that satisfies environmental regulation.

This cannot be retrofitted onto a site that’s physically unable to support compliant discharge, and it’s often discovered too late, after land is committed.

5. Cleanroom envelopes

Classified space is a constructed and qualified system: materials of construction that can be cleaned and will not shed, sealed junctions, appropriate finishes, airlocks and material and personnel flows designed to prevent cross-contamination, all classified and qualified to standard. Cleanroom construction is a specialist discipline, not a fit out.

6. Clean utilities and supporting systems

Clean steam, compressed air and nitrogen at pharmaceutical quality; chilled water; controlled temperature and cold chain warehousing with quarantine areas; QC laboratories with stability chambers and analytical instrumentation; hazardous material and solvent storage with fire protection; and data systems capable of supporting data integrity requirements. Each is a qualification burden as well as a capital item.

Why this is a different problem in Nigeria

In an established pharmaceutical cluster, most of the above is procured. Grid power is reliable; industrial water and effluent services exist plus specialist maintenance is available locally.

In Nigeria, a manufacturer must generally provide all of it. This means to make medicine, you must first become a power company, a water utility and a waste treatment operator then run all three to pharmaceutical reliability standards, permanently, alongside your actual business.

This creates three compounding problems:

It inflates capex far beyond the manufacturing plant itself. A large share of the investment goes into systems that produce no product.
It permanently inflates opex and therefore unit cost. Self-generated power, self-treated water and self‐managed effluent are structurally more expensive than purchased utilities, and that cost sits in the cost of every unit produced, forever. This is often what results in loss‐making product lines rather than achievable market prices.
It demands scarce management attention and scarce technical skills. Utility operations pull focus and expertise away from manufacturing and quality, in a market where that expertise is already the binding constraint.

There is also an efficiency trap. Utilities sized and made redundant for a single manufacturer are inherently inefficient. Redundancy means paying for capacity that is idle by design. A single tenant carries the full cost of that idle capacity alone.

What this means for your feasibility study

A Nigerian pharmaceutical feasibility study that treats infrastructure as a set of assumptions rather than a costed, engineered and qualified workstream is not bankable, and a development finance institution will identify that quickly. The systems above should be specified, costed, and risk assessed with the same rigour as the manufacturing process itself because they will determine the operating cost that decides whether the business case closes.

This is the specific problem Genesis Science Park exists to solve. An IaaS life sciences park provides the power, water, effluent and GMP‐grade infrastructure layer at park level, so that redundancy and scale are shared rather than duplicated by every tenant, and so a manufacturer can be a manufacturer rather than a utility operator.

This shifts the biggest capital expenditure to opex, reducing the financial burden by up to 70%. More so, it removes the permanent cost of idle operations, contributing to expensive market prices, from manufacturers’ balance sheets.

The primary question on plant qualification is removed as a risk factor upfront. An IaaS science park is bankable by design.

→ Express your interest and download the technical prospectus.

For manufacturers that are undertaking a feasibility study, or are operational and expanding.

FAQs

Shared infrastructure sounds efficient, but doesn’t it create one enormous single point of failure, if the park’s power or water goes down, every tenant goes down at once?

TThis is the right question to ask, and pooling a dependency does concentrate it — so the answer turns entirely on redundancy. Shared infrastructure can be more reliable, not less, precisely because a park serving many tenants can afford a level of redundancy no single manufacturer can justify alone: multiple generation sources, N+1 or N+2 configurations, duplicated critical systems whose cost is spread across the whole site. A single tenant self-provisioning almost always under builds redundancy, because over building for one plant is uneconomic. Pooling done badly is a shared failure but pooling done well is shared resilience, so the question to put to any operator is not "is it shared?" but "to what redundancy standard, and what is the guaranteed uptime?"

As a tenant I’d be handing control of my most critical systems to a landlord. In pharma I can’t outsource accountability for my own GMP. What happens to my compliance, and my liability, when the operator fails?

It is correct that you can’t outsource GMP accountability, and you should never accept a model that pretends you can. The distinction that makes this work is between the shared utility supply and your qualified systems - the park delivers power, raw and pre-treated water and effluent handling to a defined, evidenced standard, while the product contact systems inside your envelope - your purified water and water for injection loops, your cleanrooms, your process - remain yours, qualified and controlled by you. What you should demand, and what a serious operator provides, is the qualification and monitoring documentation for the shared layer so it can feed your own validation, plus service level terms that make the operator’s obligations and liabilities explicit. Dependency without documentation is the risk, dependency with contractual and qualification evidence is just supply chain management.

Doesn't sharing utilities across tenants create cross contamination risk or regulatory contagion, where one tenant’s process or one tenant's failed inspection taints the whole park?

Cross contamination is a cardinal GMP failure, so the concern is legitimate and the design answer has to be explicit. The principle is that the shared layer is the non-product contact commodity layer - bulk power, incoming water before final treatment, effluent after it leaves your process - while everything that touches product stays dedicated, segregated and individually qualified per tenant. Done properly, tenants share the supply of a utility, not the utility at the point it meets their product. On regulatory contagion, each facility holds its own licences and is inspected on its own systems, so a shared campus does not merge compliance identities. Though the operator’s standards do affect everyone, their qualification regime must be comprehensively scrutinised during due diligence.

Won’t shared utilities actually cost me more (an operator margin stacked on top) than building my own once and running it at cost?

A shared model does carry an operator return, so it is not automatically cheaper, and you should model it rather than take it on faith. But compare total cost of ownership, not the sticker. Self-provisioning means paying the full capital cost of redundant capacity you will only partly use, then carrying its operating cost, its management burden and the permanent waste of idle redundancy by yourself. Sharing spreads the redundancy capital and the utilisation across multiple tenants, so even net of an operator margin the delivered unit cost can be lower plus it converts a large upfront capital outlay into an operating cost, which materially changes your financing profile and your feasibility case. Whether it wins for you depends on utilisation and the operator’s efficiency, which is exactly what a bankable study should test rather than assume.

"GMP compliance is basically an energy load" feels like an oversimplification. Compliance is documentation, quality systems, trained people and validated processes not just air handling.

Absolutely, and it’s worth being precise. GMP compliance is overwhelmingly about quality systems, documentation, validated processes and competent people, and HVAC is a small part of the whole. The narrower, specific claim is about infrastructure: of the physical plant you must build and power to enable compliant manufacturing, environmental control is the dominant energy consumer, with cleanroom HVAC reported at roughly 36 to 67% of facility energy. So the point is not that compliance equals energy, it’s that the infrastructure layer compliance depends on is energy dominated, which, in a market where reliable power is the scarcest input, is what turns an engineering line item into the binding constraint. The documentation and quality systems remain entirely yours to build.