Bankable Solar Park Design: A Risk-Led, Front-Loaded Process

A utility-scale solar park is made insurable and bankable at the design stage, long before the first pile goes in. Lenders’ technical advisors judge the design, the equipment and the contracts; insurers judge how exposed the plant is to hail, wind, flood, fire and theft; and both want evidence that risks were found and designed out rather than discovered on site. My position is that bankable solar park design should be front-loaded and risk-led: detailed surveys feed a risk assessment, the risk assessment shapes the design, and the risk management plan and QA/QC plan are written before procurement starts.

Why bankable solar park design is settled before procurement

The International Finance Corporation (IFC) guide for utility-scale PV developers calls a full, independent technical due diligence of the design before construction good practice, and an essential requirement where financing is sought ^[1]. Its example is a 5 MWp plant built in India in 2010, where independent consultants found purlins sagging under the module load, corrosion-prone mild-steel fixing rods and a foundation type that made module alignment difficult. Extensive remedial work followed ^[1]. Each fault was a design or specification decision.

The EU-funded Solar Bankability project reached the same conclusion from data. It separates “year 0” risks, whose root cause lies in component production, design, planning or installation before commissioning, from risks that originate in operation ^[2]. Its guidelines, built on due diligence reports, O&M records and claims data, state that the most effective mitigation is prevention, followed by the earliest possible detection; preventive measures are paid for mostly from CAPEX, corrective ones from OPEX ^[2]. An IEA PVPS worked example shows the scale: potential-induced degradation on a 10 MWp plant left its 20-year profit 48% below expectations, while mitigation measures, including their cost, kept the shortfall to 5-6% ^[6].

Hence the position behind this article: put more of the project’s time and engineering effort into surveys, risk assessment and design than is usual, so that construction executes decisions instead of making them. Two limits apply. I know of no evidence for a fixed ratio of design time to construction time; the evidence supports front-loading effort into prevention. And design work has diminishing returns and does not replace quality control during execution. Where grid or permit deadlines compress the design window, that compression belongs in the risk register as a risk itself.

What lenders and insurers look for

Lenders’ technical advisors

IFC defines the scope of technical due diligence as the technology, energy production profile, design, construction risks, integration, and the technical aspects of the permits and of the EPC and O&M contracts ^[1]. For large projects, the Solar Bankability guidelines add component testing of modules, possible production audits at module and inverter suppliers, audits during installation and commissioning, and performance monitoring in operation ^[2]. They also remind investors that they carry every residual risk not covered by warranties, guarantees or insurance, so they should assess technical risk actively rather than rely on the bank’s review ^[2]. What an advisor looks for is traceability: each design choice tied to a survey result, a standard or a documented risk decision. A correct but undocumented design still generates findings.

Insurers

A bankable insurance program typically combines marine cargo, delay in start-up, construction all risks and third-party liability during construction, followed by operating all risks (including machinery breakdown), business interruption and liability in operation ^[3]. In global loss-adjuster data presented by the broker Marsh, storm and hail made up 12% of claims but 80% of claim value, while poor workmanship, theft and a grid-related issue made up 85% of occurrences at relatively low value ^[3]. kWh Analytics’ US-weighted loss database shows hail at 73% of financial losses from only 6% of incidents, with fire the second most costly peril and significantly driven by overgrown vegetation ^[4]. Its 2026 assessment adds that only 4% of PV fire loss events occurred in high wildfire-risk areas, while 84% were equipment-driven brushfires ^[7].

For Central Europe, hail deserves attention without exaggeration. A study by the European Severe Storms Laboratory (ESSL), Adam Mickiewicz University and Munich Re found that severe hail is increasing more rapidly in Europe than anywhere else, although Europe’s hotspots cluster near mountain ranges in north-eastern Spain, south-western France and northern Italy ^[8]. Hungary is not on that list, but the region is exposed. In 2023, Europe’s third record-breaking hail season in a row, a single July day brought severe hailstorms to Slovenia, Croatia, Serbia, Slovakia and Romania, all neighbors of Hungary, and a derecho whose damage path of almost 1,000 km crossed southern Austria, Slovenia, Croatia and Serbia ^[9]. For a plant expected to run 30 years or more in the Pannonian Basin, I treat hail and convective wind as design loads, not only as insurance lines. The site fixes the hazard; the design largely fixes the vulnerability, through module glass, structure, layout, vegetation strategy and security.

Surveys that de-risk the design

Survey scope is where front-loading starts. The minimum set for a utility-scale site:

• Topographic survey. IFC expects a detailed site plan with topographic contours, access routes and civil works requirements ^[1]. Slopes drive structure type, grading, row layout and surface water paths.
• Geotechnical investigation. IFC recommends boreholes or trial pits at regular intervals, with sampling and in-situ testing usually to 2.5-3 m, assessing groundwater level, soil resistivity, load-bearing properties, obstructions, and the suitability and drivability of the foundation type ^[1]. Pile testing and corrosion are covered in a separate article on pile foundations in corrosive soil.
• Flood and hydrology. EU member states must produce and update flood hazard and flood risk maps under the Floods Directive (2007/60/EC) ^[10], the natural starting point for a site study. Flooding damages electrical equipment at or near ground level and raises erosion risk around structures and foundations ^[1], so the output should be design levels for inverter and transformer platforms and cable routes. Drainage design is the subject of another article.
• Grid studies. IFC warns that a neglected grid connection study can lead to unforeseen costs that seriously damage project viability ^[1].
• Environmental and social constraints. Permits for forestry and endangered species, and social factors such as cultural heritage and archaeological sites, can impose conditions ^[1] that the layout and schedule must reflect before tender.
• Climate hazards. Site wind, snow and hail data, to set the loads and the hail criterion.

From the EPC side of more than 300 MWp of Hungarian solar projects, my view is that a survey gap never disappears. It returns during construction as pile refusal, a redesign, a flooded cable trench or a variation claim, when it is most expensive to resolve. The Solar Bankability risk report points the same way: typical installation mistakes stem mostly from lack of know-how, too little installation time and wrong planning ^[11].

From risk register to design basis

Building the risk register

ISO 31000:2018 provides generic risk management guidelines applicable to any activity and any type of risk ^[12]. For PV, the Solar Bankability guidelines use four steps: identification, assessment, management through mitigation and transfer, and controlling ^[2]. For identification, the project’s risk matrix is a useful template, sorting technical risks by phase (product testing, planning, transport and installation, operation, decommissioning) and by component, from modules, inverters and mounting structures to cabling, earthing and lightning protection, the transformer station and environmental influences ^[11].

For assessment, the Cost Priority Number method is a cost-based failure mode and effects analysis that expresses each risk in EUR per kWp or EUR per kWp per year ^{[2, 11]}, calibrated on more than one million failure cases from 746 plants totaling about 422 MWp ^[2]. Even without the full method, scoring risks in money rather than colors makes the trade-off between mitigation cost and expected loss explicit. Each entry should carry an owner, a treatment (avoid, reduce, transfer or accept), the design element or contract clause that implements it, and the residual risk.

Writing the design basis

The design-related treatments go into one controlled design basis document, which should fix at least:

• the design life;
• the load standards, including EN 1991-1-3 for snow and EN 1991-1-4 for wind ^[13], with the applicable national annex;
• the hail criterion for modules and, for trackers, the stow strategy;
• flood design levels;
• the corrosion class and durability approach for foundations;
• the earthing, lightning and surge protection concept;
• the fire, vegetation and security concepts;
• the energy yield assumptions and their uncertainties.

Input quality matters: yield simulations built on little site information carry the highest uncertainty and make a project less attractive to invest in ^[2]. The Solar Bankability risk report also lists the studies that make an installation reliable: a static study of the mounting structure combined with the suitability of the site, external lightning and surge protection, the earthing system with step and touch voltage protection, and the electrical installation study ^[11]. One further discipline: state safety margins openly. A margin hidden inside a supplier’s calculation cannot be checked, and one applied twice by two parties wastes steel.

Reviewing the design

The design review should be staged (concept, basic, detailed), run against the risk register, and close every comment in writing before the next stage. The lender’s technical advisor then reviews a design that has already been challenged.

Equipment quality and supplier selection

Module certification is a floor, not a selection tool. In Kiwa PVEL’s 2026 Scorecard, 87% of manufacturers and 61% of BOMs had at least one failure in its extended test program; breakage in the mechanical stress and hail stress sequences remained the leading failure mode, and 26% of BOMs in mechanical stress testing had at least one broken module ^[5]. Two BOMs from one manufacturer lost 14.1% and 3.6% of power in the same UV test ^[5]. The unit of quality is the BOM, not the brand.

Hail needs particular care. The minimum IEC 61215 hail test uses 25 mm ice balls at 23 m/s, which PVEL calls insufficient proof of hail resistance in hail-prone regions ^[14]. Kiwa PVEL’s top-performer threshold for its hail sequence is no glass breakage with hail of 40 mm or more ^[5]. SolarPower Europe’s 2026 EPC guidelines likewise highlight hail and recommend stronger design and procurement measures ^[15].

Practical selection criteria follow:

1. BOM-specific evidence: extended test reports for the exact BOM offered, from factory-witnessed samples, as Kiwa PVEL’s Scorecard also requires ^[5].
2. A BOM lock in the supply contract, with re-approval for any material change.
3. Production audits at module and inverter factories, part of the Solar Bankability due diligence scope for large projects ^[2].
4. Packaging, transport and handling requirements, since module condition on delivery is often unclear and handling requirements often poorly specified ^[11].
5. Spare parts and lead times for replacement equipment, which brokers list among key solar risks ^[3].
6. Warranty terms and the counterparty behind them.

Bankable supplier lists are a reasonable first filter, but in my view they describe a manufacturer’s commercial standing, not the quality of the batch a project receives.

Write the risk management plan and QA/QC plan before procurement

The risk management plan turns the register into commitments. The Solar Bankability guidelines recommend transferring risks through contracts, guarantees, warranties and insurance to the parties best able to control them ^[2]. EPC warranties for defects and workmanship typically run one to two years, sometimes five, and an on-site audit before expiry is highly recommended; insurance transfers part of the remaining risk, but deductibles stay with the owner ^[2]. The plan should state, for each major risk, who carries it during construction, during the defects period and afterward, and what evidence closes it.

The QA/QC plan is the register translated into checks. Every high-ranked risk needs an inspection or test with an acceptance criterion and a hold or witness point: factory and pre-shipment inspection, incoming inspection on site, installation checks, commissioning tests and complete as-built documentation, which the Solar Bankability risk report treats as necessary for safe maintenance and troubleshooting ^[11]. Executing the plan is the subject of a separate article on QA/QC from factory to commissioning. The point here is timing: a QA/QC plan written after the purchase orders cannot add a factory audit or BOM test the contract never required.

A design-phase de-risking roadmap

This is the sequence I recommend. Each step has an output and a gate to pass before the next begins.

1. Surveys: topographic, geotechnical, flood, grid, environmental and climate-hazard studies. Gate: design inputs available, gaps logged as risks.
2. Risk register: risks by phase and component, scored (ideally in EUR per kWp), each with owner and treatment. Gate: reviewed with the owner and, where possible, the lender’s advisor and the insurer’s risk engineer.
3. Design basis: design life, loads, hail and flood criteria, corrosion, earthing and lightning, fire, security and yield assumptions. Gate: frozen under change control.
4. Design review: staged independent review against the register. Gate: all comments closed in writing.
5. Procurement specifications: BOM evidence and lock, factory audits, packaging and handling, spares, warranties. Gate: every clause traceable to the register or design basis.
6. Risk management plan and QA/QC plan: risk allocation across contracts and insurance; inspection and test plans with hold points. Gate: both issued before the first purchase order.

The table maps the main risks to design responses and the evidence a lender’s advisor or insurer will expect.

Conclusion

Insurability and bankability are not certificates added at financial close. They are the visible result of decisions made during surveys and design: what was measured, which risks were ranked, which margins were chosen, which BOM was locked and which checks went into contracts. Front-loading that work costs time and fees before procurement, and the evidence suggests it is the cheapest point to buy down risk. The test is simple: when the lender’s advisor or the insurer’s risk engineer asks why a choice was made, the answer should already be in the risk register.