What Running a Remote Resort on Diesel Actually Costs – and Where Solar and Storage Change the Equation

A resort that generates its own power usually starts with one number: the diesel price at the port, multiplied by the litres it burns each month. That number is real, it arrives on an invoice every month, and it is almost never the actual cost of the electricity.

The gap between the two is where most of the money hides. Fuel that has to be shipped, barged and stored before it ever reaches the generator. A generator that spends much of the day running well below the load it was sized for. A maintenance programme whose cost climbs with every kilometre between the site and the nearest technician. All three are real expenses — they simply do not appear on the same line as the fuel bill, which is exactly why they get underestimated when the decision is being made.

This is not a theoretical problem. Island resorts that have already made the change publish their numbers, and the range is wide: from roughly 117,000 litres of diesel displaced per year at a small island project to nearly a million litres at a large resort. What separates those outcomes is not the technology. It is the shape of the site’s own load curve and the price it actually pays for fuel.

This article takes the diesel bill apart into the three parts that drive it, looks at what documented resort projects achieved, and sets out what determines whether the same arithmetic works at your site.

Two horizontal diesel storage tanks inside a bunded containment wall beside an open-type diesel generator set on a concrete plinth at an island resort, with the sea visible beyond

The Diesel Price Is Not the Cost

The price quoted at the port is the starting point, not the delivered figure. Between the terminal and the generator there is a chain of costs that every remote site pays in some form: sea freight, barge or lighter transfer where the harbour cannot take the supply vessel, road transport on the final leg, on-site storage tanks, spill containment, and the security needed to keep a valuable and easily resold commodity in place.

Industry analyses of island fuel supply consistently put the delivered premium at 30–50% above mainland prices. On top of that sits volatility. The diesel price a resort pays is exposed to global crude markets it has no influence over, and island supply chains tend to amplify movements rather than absorb them.

That combination has already forced the issue in several markets. In Fiji’s Yasawa group, an 80% surge in diesel prices pushed resorts toward solar, with one operator projecting annual fuel savings of around FJ$850,000. That is the pattern: the switch rarely happens because a resort decided to become greener. It happens because a fuel price moved and the operating margin could not absorb it.

The economics of remote fuel logistics are covered in more detail in our earlier analysis of why diesel delivery costs balloon at remote sites — the short version is that the delivered price, not the quoted one, is the number the whole calculation has to run on.

Why the Cost per kWh Climbs as the Load Falls

A resort’s load is not a factory’s. It rises in the morning as kitchens, pumps and air conditioning come online, holds through the middle of the day, climbs again in the evening when lighting, restaurants and guest rooms peak, then falls away to a small overnight base load. The daily swing is large, and the peak is what the generator gets sized for.

That sizing decision has a consequence that shows up every night.

A diesel engine’s fuel consumption per kilowatt-hour is lowest in a band roughly between 70% and 85% of its rated output. Below that, the fixed losses — cooling fans, oil and water pumps, friction, excitation — stay roughly constant while the useful output falls away, so each kilowatt-hour delivered costs more fuel than the last. At 25% load, a generator can burn more than 40% more fuel per kWh than it does at full load.

At a resort with several generators installed for redundancy, the overnight picture is worse than the arithmetic suggests. If four units are installed, one may be carrying the entire overnight base load alone — running at a small fraction of its rating, for hours, at its worst efficiency point. The fuel-versus-load relationship is what turns a modest overnight load into a disproportionately expensive one.

There is a second cost that does not appear on the fuel invoice at all: running hours. Every hour the engine turns brings the next service interval closer, and every hour below the recommended minimum load contributes to wet stacking — unburned fuel and carbon accumulating in the exhaust path, the turbocharger and the injectors. A generator that runs 8,000 hours a year and a generator that runs 2,000 hours a year are not the same asset at the end of it, even if they were identical when installed.

Aerial view of a low-rise island resort with solar modules across the villa and restaurant roofs, a jetty and turquoise lagoon, and a diesel generator and fuel tank compound at one edge of the island

What Maintenance Looks Like When the Site Is an Island

On the mainland, generator maintenance is a scheduled visit. On an island it is a flight, a transfer, a technician’s time and often an overnight stay — and the cost of those logistics does not scale down with the size of the job.

The maintenance calendar itself is driven by running hours rather than the calendar. Routine service points commonly fall around 250, 500 and 1,000 operating hours: oil and filters at the shorter intervals, air, fuel and oil filtration as the interval lengthens. A major overhaul typically arrives around 15,000 equivalent operating hours, and it is a different order of expense — a planned shutdown, specialist labour, and a set of parts that has to be shipped to site in advance.

Published cost benchmarks vary widely by context and size. HOMER’s knowledge base, drawing on high-cost Alaska conditions, cites operation and maintenance on the order of $14 per operating hour for units in the 400–800 kW range. EPRI data for a 2 MW standby unit puts scheduled maintenance materials and labour near $0.01/kWh, excluding fuel and operating labour. The spread between those figures is itself the point: the further and more difficult the site, the more of the total is logistics rather than parts.

Availability matters too. Diesel generators are typically assumed to run at around 95% availability, and when a failure happens at a remote site it commonly takes several working days to resolve once travel is included. That figure feeds directly into how much redundancy a resort carries — which in turn means more installed capacity, more units, and more of them running at low load.

What the Documented Resort Cases Actually Show

Several island resorts and island communities have published what their hybrid systems delivered. The figures below are as reported by the project developers, operators or the media covering them; configurations differ, so they are a range of outcomes rather than a like-for-like comparison.

Project Configuration Annual diesel displaced
Large resort, Maldives 2 MWp floating PV + 3 MWh battery 900,000–1,000,000 litres (generators off 16–18 hrs/day)
Resort, Dhaalu Atoll, Maldives 1,074.57 kWp PV + 1,064 kWh BESS ~800,000 litres (~2,120 t CO₂, ~€560,000 operating savings; 1 of 4 generators running)
Hilton Maldives Amingiri 770 kWp rooftop PV + 540 kWh battery ~321,000 litres (~848 t CO₂)
Lord Howe Island, Australia 1.3 MW solar + 3.7 MWh battery with existing diesel 349,082 litres (year 1); 360,236 litres (year 2); ~$1.55M fuel savings over two years
Baa Maalhos, Maldives 396 kWp solar–diesel–battery + 716 kWh BESS 185,837 litres
Innamaadhoo, Maldives 350 kWp PV + 618 kWh BESS 164,672 litres (generators off ~6 hrs/day)
Baa Kendhoo, Maldives 249 kWp hybrid + 344 kWh BESS 117,181 litres

Read together, three things stand out.

The savings scale with the size of the array and the battery, not with the sophistication of the controls. The projects displacing the most fuel are simply the ones with the most solar and the most storage behind it.

The mechanism is always the same: the battery lets the generators stop. The reported 16–18 hours a day at the largest project, and roughly 6 hours at the smallest, describe the same thing at different scales — diesel is no longer the source that must always be online.

The outcome at a given site is not predictable from someone else’s numbers. Two resorts with identical diesel prices and identical solar resources can land a factor of several apart purely because of how their load curves are shaped.

For resorts still deciding what to build, the design questions — load profile, architecture, component sizing, noise and sight lines, redundancy — are covered in our guide to designing a reliable microgrid for resorts and remote hotels.

Which Lines of the Bill Solar and Storage Actually Attack

A hybrid system does not reduce every cost on the list, and being precise about which ones it does reduce is what makes the payback estimate honest.

What it attacks directly. Daytime fuel, because solar carries load that the generator would otherwise supply. Low-load running hours, because a battery can absorb the overnight base load that a large generator handles badly — and when the generator does start, the battery lets it run at a healthier share of its rating instead of idling. Total running hours, which push out the service intervals and postpone the overhaul.

What it reduces but does not remove. Maintenance does not disappear. The generator still needs servicing, still needs an overhaul eventually, and the site still needs a technician. What changes is how often, because the intervals are counted in running hours and those hours fall sharply. Some island projects report generators running a few hours a day instead of continuously.

What it does not change. The generator itself stays. It remains the resource that covers a week of heavy cloud, a period of unusually high demand, or a failure elsewhere in the system. Every documented resort example above keeps diesel installed; the difference is that diesel moves from being the source that always runs to the source that runs when it is needed. Battery storage absorbs the short-term swings that would otherwise force the generator to follow every fluctuation, which is why a battery energy storage system is usually sized for the evening and overnight window rather than for the whole day.

Getting that balance right — how much solar, how much battery, how much diesel, and how the three are dispatched against each other — is the core of what a solar-battery-diesel hybrid microgrid is designed to do. It is also where the difference between a system that pays back and one that does not is decided.

White battery energy storage containers with cooling units, a switchgear and transformer enclosure, and an open-type diesel generator set with exhaust silencer standing in one gravel compound at a remote island site

What Decides Whether It Pays Back at Your Site

The documented cases give a range, not a promise. Five variables explain most of the variation between a strong outcome and a weak one.

The delivered diesel price. This is the single largest lever. A site paying $0.45/kWh for diesel-generated electricity has far more room to work with than one paying $0.20. Published benchmarks put diesel-only costs on islands in the region of $0.30–0.45/kWh under normal conditions, and higher when fuel prices spike, against roughly $0.10–0.18/kWh reported for solar-plus-storage hybrids in comparable settings.

The shape of the load curve. A resort with a pronounced day–night swing, a heavy evening peak and a low overnight base gives solar and storage much more to work with than a site with a flat, round-the-clock industrial load. The wider the gap between peak and base, the more hours a battery can carry alone.

The solar resource. Seasonal cloud cover and day length set how much of the load solar can realistically offset. Two islands at similar latitudes can differ substantially once the wet season is accounted for.

The battery’s autonomy. This is what decides how long the generators stay off — and it is the variable that most directly converts into litres saved. It is also the most expensive component, so it is the one that has to be sized against the load curve rather than against a target renewable percentage.

The condition of what is already installed. A resort with newer, well-maintained generators is a different proposition from one running units that are approaching overhaul. In the second case, the question is not only whether to add solar and storage, but whether to plan the replacement differently.

It is also worth being clear about what the numbers above are not. They are project disclosures at specific sites, not a forecast for yours. Island microgrid studies from IRENA and UNEP put the LCOE of fully diesel-based island grids at 34.7–48.3 US cents/kWh, with hybridisation reducing that by roughly 9–20% — a much more conservative range than the headline resort figures, and a reminder that the outcome depends heavily on fuel price assumptions and financing terms.

YURI POWER designs and delivers solar-battery-diesel hybrid power systems for island and remote sites, working from the site’s actual load profile, fuel logistics and operating requirements rather than from a standard configuration. If you are running a resort or remote facility on diesel and want to understand what the numbers would look like at your site, our team can work through the load and cost picture with you.

FAQ

Q: What does it actually cost to run a resort on diesel?

A: The delivered fuel price is the starting point, but the electricity cost is higher than fuel alone implies. Published benchmarks put diesel-only generation on islands in the region of $0.30–0.45/kWh under normal conditions, rising when fuel prices spike. The figure depends on the delivered diesel price, how the load curve is shaped, and how many hours the generators spend at low load.

Q: How much diesel can solar and storage realistically replace?

A: Documented resort projects range from around 117,000 litres a year at smaller installations to nearly a million litres at large ones. The variation is driven mostly by array size, battery capacity and the site’s own load curve — not by the control technology. Any figure quoted for your site should be derived from your load data, not from someone else’s result.

Q: Do we still need the diesel generators if we add solar and storage?

A: Yes. Every documented resort example keeps its generators installed. What changes is their role: diesel stops being the source that runs continuously and becomes the source that runs during extended bad weather, peak demand and emergencies. Removing it entirely is a different design decision, and one that usually requires a much larger battery.

Q: What is a typical payback period?

A: It varies too widely to quote responsibly as a single number. Payback depends mainly on the delivered diesel price, which projects show is often the dominant factor, and on how many running hours the battery allows the generators to avoid. Several resort projects report payback periods under five years, but those are site-specific results against their own fuel prices and load profiles.

Q: Can the system carry the resort at night?

A: That is precisely what the battery is sized for. The evening and overnight window is where a large generator runs at its worst efficiency and where storage has the most to offer — provided the battery is sized against the actual overnight load rather than against a renewable-energy percentage target.

Q: What happens during a week of heavy cloud?

A: The generators run. This is the case they are kept for, and it is why the diesel system stays in place rather than being removed. A hybrid system is designed so that poor solar days increase generator running hours rather than threatening supply, and the battery covers the gap on shorter cloudy periods.

Q: Can we add solar and storage to the generators we already have?

A: In most cases, yes — the existing generators can remain as the backup layer while solar and battery are added alongside them. Whether that is the better route depends on the age and condition of the installed units, their suitability for cycling, and how their controls integrate with the new system.

YURI POWER Engineering Team · September 18, 2026