Heat source modernisation and replacement for businesses – how to compare the available options?

Replacing a heat source should rarely start with the question: “Which boiler, heat pump or CHP system should we choose?” First, you need to determine how much heat the plant actually needs, when it needs it, at what temperature level and with what load variability. Only then can you make a sound comparison of modernising the existing source, electrification, a heat pump, CHP, waste heat utilisation or a hybrid system.

In practice, the biggest mistake often occurs before technologies are even compared: the new source is sized to today’s consumption even though the building, ventilation, process, heat recovery or system parameters are about to change. The equipment can then be sized correctly, but for the wrong load.

In brief: where should heat source replacement begin?

• Do not size capacity solely to the current peak. First check whether demand will fall after modernising the building, systems or process.
• Separate base, variable and peak load. One technology may be optimal for a load that occurs for several thousand hours per year, while another may be better for short periods of maximum demand.
• Group heat users by required temperature. Space heating, hot water, a process bath, drying and process steam do not all have to be supplied from one source at the same temperature or pressure level.
• Compare the cost of useful heat and the cost of the entire system, not just the fuel price. Seasonal efficiency, network charges, contracted capacity, maintenance, connection costs, distribution and the source operating strategy also matter.
• Do not assume that one technology must serve the whole plant. In many facilities, a cascade or hybrid arrangement is more rational: one source for the base load, another for peaks, while heat recovery covers part of the demand whenever it is available.

Heat source replacement starts with the demand profile

Annual fuel or heat consumption shows how much energy the plant used, but not how much capacity it needs in individual hours. Yet it is the load profile that determines whether the source will operate close to an efficient operating point for most of the year, or whether it will run only for short periods and remain heavily underloaded much of the time.

Before choosing a technology, it is therefore worth building an hourly profile or – if data are limited – at least profiles for representative days and seasons. A sound analysis should distinguish between:

• base load – demand that occurs for a large part of the year or production time;
• variable load – load that depends on production, weather, work shifts or the number of active heat users;
• peak load – the highest load, usually occurring for a relatively short time;
• critical loads – heat users that must remain supplied during failures, partial plant shutdowns or maintenance.

Only this breakdown makes it possible to determine whether a single source, a cascade of several units or a combination of technologies makes sense. A source with high capital cost but low operating cost may be suitable for the base load. A simpler unit with a higher variable cost may be rational as a peak source if it operates only for tens or hundreds of hours per year.

What data are needed?

The most useful input data set includes:

• gas, fuel, district heat and electricity consumption for at least 12 months, and preferably 24 months where seasonality is strong;
• hourly data or shorter intervals from meters, BMS/EMS, heat meters or measurement systems, where available;
• supply and return temperatures, steam pressure, flows and actual operating parameters of heat users;
• production schedules, work shifts, shutdowns, start-ups and seasonality;
• a list of heat users with their required temperature and capacity;
• parameters of the existing boiler house or other source: capacities, efficiencies, modulation, number of units, controls and operating history;
• condition of the distribution network: insulation, route lengths, temperatures, losses, hydraulics and throttling points;
• available electrical capacity, transformer ratings, connection conditions, gas, district heat and other available utilities;
• potential waste heat streams: medium, temperature, flow, availability period, contaminants and distance to the heat user;
• planned production changes, building energy retrofits, ventilation replacement, heat recovery or process modifications.

If a plant has no heat meters, the profile does not always have to be guessed. In many systems it can be reconstructed from a combination of fuel data, temperature and flow measurements, equipment operating time and short-term measurements taken under representative conditions.

Heat-user temperature is just as important as capacity

Two plants with the same thermal capacity may require completely different solutions. One may mainly need water at 50–70°C for space heating and washing, while another requires process steam or high process temperatures. A simple comparison of “MW to be replaced” therefore misses a crucial part of the problem.

It is worth preparing a temperature-level map and checking whether all heat users really have to be supplied at the highest temperature or pressure level. An unnecessarily high hot-water temperature or steam pressure can increase losses and rule out technologies that would be much easier to apply at lower temperature or pressure levels.

This is particularly important for electrification and heat pumps. According to the IEA Heat Pump Monitor 2026, industrial heat pumps below around 120°C have already entered commercial operation, while solutions below around 160°C are starting to enter commercial operation; higher temperature levels remain less mature. This does not mean, however, that one specific temperature is the decision boundary. Source temperature, the required temperature lift, refrigerant, process conditions and operating hours also matter.

Auditor’s comment: before concluding that a plant “needs 120°C”, it is worth checking which heat user actually requires it. Sometimes the whole network is operated at a high temperature or pressure solely because of one process. Separating the circuits can open up completely different options for the rest of the load.

First check how demand will change after efficiency improvements

Sizing a new source to the current capacity only makes sense if the post-investment load remains similar. In practice, this is often not the case.

Before ordering the source, it is therefore worth calculating the impact of measures that may change the heat balance, including:

• insulating the building envelope, replacing gates and reducing infiltration;
• changing heating and ventilation schedules;
• optimising temperature settings and controls;
• reducing losses from pipelines, valves and tanks;
• reducing heating-water temperature or steam pressure where the heat users allow it;
• modernising heat exchangers and the production process;
• recovering heat from refrigeration, compressors, flue gases, wastewater, products or other process streams;
• changes in production, floor area, operating hours or the number of shifts.

In buildings, the interactions between the envelope, ventilation, controls and heating are especially important. The scope of the energy retrofit and the parameters of the new source should therefore be analysed together, not as two independent projects.

It is useful here to distinguish engineering practice from legal requirements. Article 3 of Directive (EU) 2023/1791 requires Member States to ensure that energy-efficiency solutions are assessed in planning, policy and major investment decisions worth more than EUR 100 million, or more than EUR 175 million for transport infrastructure projects. This is not a direct rule requiring every company to follow an identical sequence for every boiler-house replacement. In this guide, the principle “first reduce and define the need, then select the source” is primarily an engineering rule.

How should the available heat-source options be compared?

There is no single list of technologies that is right for every business. The U.S. Department of Energy notes that the diversity of industrial processes requires a portfolio of solutions selected for the specific application, economics, location and other conditions. In practice, the comparison should include at least a baseline option and several realistic alternatives.

1. Modernising the existing source

A “heat source replacement” does not always mean changing the technology completely. Sometimes the most rational option is to modernise the existing boiler house or source: replace the burner, improve controls, use a cascade of smaller units, install an economiser, change operating parameters, upgrade automation or reduce distribution losses.

The efficiency of the source itself must be distinguished from the efficiency of the whole system. A new boiler will not solve the problem if heat is later lost through poorly insulated pipelines, the system operates at an unnecessarily high temperature or pressure level, or the source is oversized and runs at low load for most of the time.

2. Electrification

Electrification may mean an electric boiler, heating elements, resistance, induction or microwave heating, or another technology selected directly for the process. Its advantages may include high point-of-use efficiency and precise control. That alone, however, is not enough for an economic assessment.

Before choosing an electric option, check the available connection capacity, transformer loading, impact on maximum demand, electricity price profile, network charges and the possibility of flexible operation. In a plant with limited spare capacity, full electrification may require major infrastructure upgrades, while partial electrification of the base or peak load may be possible with much less intervention.

3. Heat pump

A heat pump is particularly attractive when the plant has both a useful heat source at the highest practicable temperature and a heat user that does not require a very large temperature lift. Potential sources include refrigeration heat, process water, wastewater, process air or another waste stream.

A single COP value from a catalogue should not be used for comparison. The IEA points out that heat-pump operating costs depend heavily on energy prices, while IEA HPT Annex 58 emphasises the importance of defining and guaranteeing COP for specific operating conditions. The greater the required temperature lift, the more important the actual equipment and integration characteristics become.

A heat pump also does not have to cover the peak load. A system in which the heat pump operates for many hours at base load while an existing or new peak source runs only periodically may be technically and economically better than trying to replace the entire boiler house with a single unit.

4. Waste heat utilisation

Before building a new source, it is worth checking whether some of the required heat is already being generated in the plant. The DOE identifies typical sources including hot exhaust gases, cooling water, hot equipment surfaces and heated products.

The first priority should be to look for direct use of heat at an appropriate temperature level. If the temperature is too low, one option may be to raise it with a heat pump. In both cases, timing is crucial: waste heat has value only when a heat user needs it at the same time, or when the heat can be stored in a technically and economically reasonable way.

5. Combined heat and power (CHP)

CHP produces electricity and useful heat simultaneously. It should therefore not be compared with a boiler solely in terms of thermal efficiency. The U.S. EPA notes that CHP is best suited to facilities with sustained, coincident demand for electricity and heat.

The analysis should therefore determine how much heat will actually be useful in each period, what the electricity profile looks like, how many hours the unit can operate and what happens when heat demand falls. A detailed assessment of the technology and its economics should be performed separately; in this guide, CHP is one of the heat-supply options considered in the system balance, not the default choice.

Why is the lowest fuel price not enough?

A comparison such as “gas costs X, electricity Y, so we choose the cheaper energy carrier” is too simplistic. A company does not buy fuel simply to consume it – it needs useful heat at the required temperature or pressure level and at the required time.

As a first approximation, the variable cost of useful heat can be expressed as:

unit cost of useful heat ≈ unit cost of input energy / seasonal efficiency + variable auxiliary costs per unit of useful heat

For a heat pump, the denominator is replaced by the actual COP/SCOP under the relevant operating conditions. For CHP, such a simple ratio is not enough: the simultaneous production of electricity and useful heat must be taken into account and the entire system compared with the reference option.

This is still not a complete analysis. TCO or a cash-flow model should also include:

• CAPEX for equipment and infrastructure;
• utility connections, transformer, gas pressure-regulating station, stack, fuel storage or other auxiliary equipment;
• fixed and variable energy and capacity charges;
• energy used by pumps, fans and auxiliary equipment;
• maintenance, inspections, parts, consumables and overhauls;
• availability, planned shutdowns and reserve cost;
• service life and major replacements during the analysis period;
• financing cost;
• sensitivity of the result to energy prices and changes in the production profile.

The lowest fuel price today may therefore lose to a solution that uses less input energy, matches the load better or has lower maintenance costs. Conversely, highly efficient equipment may still be uneconomic if it uses an expensive energy carrier, requires a costly connection or operates only for a short time.

Why does the highest catalogue efficiency not determine the best solution either?

Catalogue efficiency describes equipment under specified conditions. A plant, however, operates throughout the year at many different operating points.

A seasonal analysis should include, among other things:

• part-load operation and modulation range;
• number of starts and standby periods;
• actual supply and return temperatures;
• heat-source and heat-sink temperatures for a heat pump;
• variation in production and weather;
• transmission and storage losses;
• energy used by auxiliary equipment;
• heat-exchanger fouling, process conditions and plant availability;
• degree of useful heat utilisation in CHP.

In a sound comparison, the options are therefore calculated against the same load profile – hourly, by load bands or at least for representative periods. Only then is it clear which technology covers the base load, which covers the peak, and how much energy each will actually deliver.

Example: why the sequence of actions changes the result

Consider a model plant whose current system has:

• peak capacity of 2.4 MW;
• base load of about 0.8 MW during production hours;
• annual heat demand of 8,000 MWh;
• one process circuit operating at 80°C;
• an available waste heat stream of 0.45 MW at 55°C for 3,500 hours per year.

Before selecting a new source, however, demand-side measures were implemented: distribution losses were reduced, schedules were improved, building demand was lowered and one heat user was modified so that it could operate at 65°C. After these changes, the model heat demand is 6,800 MWh/year and the peak falls to 1.9 MW.

If the source had been ordered earlier on a “1:1” basis, it would have been sized at 2.4 MW instead of 1.9 MW. The 0.5 MW difference is about 21% of the original peak capacity. This does not automatically mean 21% lower investment cost, but it shows how early sizing can lead to oversizing.

The same applies to waste heat. A capacity of 0.45 MW multiplied by 3,500 hours gives 1,575 MWh of potentially available energy. If, however, the relevant heat user operates simultaneously for only 2,200 hours, then even before heat-exchanger efficiency and system constraints are taken into account, the maximum heat that can be used by that heat user is 990 MWh. Waste heat capacity is therefore not the same as useful annual energy.

Only against the updated balance should options be compared, for example:

• modernising a source sized to the new peak;
• a heat pump covering a substantial part of the base load together with a peak source;
• direct heat recovery followed by raising the temperature of part of the stream with a heat pump;
• electrifying part of the load where available electrical capacity permits;
• CHP sized to sustained, simultaneous demand for heat and electricity;
• a hybrid system combining several of the above solutions.

This example does not identify a winning technology. It shows the sequence in which the comparison should be built so that every technology is assessed against the same actual demand.

How to build a decision matrix for a heat source

Cost is important, but it should not be the only column. The same set of criteria should be assessed for every option:

Criterion weights should reflect the needs of the specific plant. For a continuous process, availability and redundancy may be more important than the shortest simple payback period. For a facility with highly variable production, modulation and the ability to expand in stages may deserve greater weight.

The most common mistakes when changing a heat source

1. Sizing capacity to historical demand despite planned building energy retrofits, heat recovery or process changes.
2. Comparing fuel prices only instead of the cost of useful heat and TCO.
3. Comparing catalogue efficiencies without bringing the equipment to common operating conditions.
4. Treating peak capacity as the capacity required throughout the year.
5. Supplying all heat users at the highest required temperature or pressure level even though some could operate at lower temperature or pressure levels.
6. Ignoring the distribution network. A new source will not eliminate losses from uninsulated pipelines or hydraulic problems.
7. Counting all waste heat as savings without checking simultaneity between the source and the heat user.
8. Sizing CHP without a coincident heat and electricity profile.
9. Ignoring electrical capacity and infrastructure in electrification projects.
10. No scenario for failure, maintenance and production growth. The cheapest option without reserve capacity may be unacceptable for a critical process.

How to carry out the analysis step by step

1. Define the baseline. Collect energy consumption and the parameters of the source, network and heat users.
2. Build the load profile. Separate base, variable and peak load, as well as critical heat users.
3. Group heat users by temperature. Check which ones genuinely require the highest temperature or pressure level.
4. Calculate the post-efficiency state. Include the building, ventilation, controls, process, insulation and heat recovery.
5. Define the reference option and realistic alternatives. Do not compare technologies that are ruled out from the outset by infrastructure or process parameters.
6. Calculate all options against the same profile. Use seasonal efficiencies and actual operating points, not just nominal parameters.
7. Compare TCO/NPV and risks. Perform sensitivity analysis for energy prices, operating hours and production changes.
8. Check whether the project can be phased. Sometimes the most rational route is to reduce demand today, implement a base-load source in the first stage and retain a flexible peak source until the next modernisation.

If the investment covers several interdependent systems, the scope of the analysis should reflect the investment objective and the available data. Energy Trend’s overview of energy audits and measurements presents the available diagnostic routes; the role of the pre-investment heat-source analysis should be to build options against a common baseline and show what actually changes the plant’s consumption, capacity and costs.

Conclusion: choose the technology at the end, not at the beginning

A good heat-source change analysis starts with the heat users. First determine how much heat is required, at what times, at what temperature level and how the profile will change after planned efficiency measures. Only then should equipment be compared.

In some plants, modernising the existing source will be the best option. In others, electrification, a heat pump, CHP or waste heat utilisation will be justified. Often, however, the best solution is not a single technology but a system in which different sources serve the parts of the profile to which they are best suited.

If you need to compare technical and economic options using plant data before investing, contact Energy Trend. It is worth carrying out the analysis before sending requests for quotations for equipment – this way the specification is based on the right demand, not the other way around.

FAQ

Should a new heat source have the same capacity as the old one?

Not automatically. First verify the actual peak capacity and the impact of building energy retrofits, process changes, controls, loss reduction and heat recovery. New capacity should be sized to the target demand, not solely to historical demand.

Can a heat pump fully replace an industrial boiler house?

In some applications it can, but this depends primarily on the required delivery temperature, the available heat source, required capacity, operating hours and electrical infrastructure. In many plants it is also worth comparing a hybrid option in which the heat pump covers the base load and another source operates at peak.

Is CHP an alternative to a boiler?

It can be one option if the plant has simultaneous and sufficiently sustained demand for heat and electricity. It should not be sized solely on the basis of annual gas consumption or peak capacity.

Should heat recovery be analysed before selecting a new heat source?

Yes, if the plant has significant waste heat streams. Temperature, availability period, overlap with demand, distance and integration options must be checked. Only useful heat that actually displaces heat production by the source should reduce the heat demand assigned to that source.

What is a better criterion than fuel price?

The cost of delivering useful heat and the total cost of ownership of the system. This includes not only fuel or electricity but also seasonal efficiency, charges, CAPEX, maintenance, infrastructure, availability and price risks.

Sources

• U.S. Department of Energy – Finding Efficiencies in Process Heat
• U.S. Department of Energy – Waste Heat Recovery Basics
• International Energy Agency – Heat Pump Monitor 2026
• IEA Heat Pumping Technologies TCP – Annex 58, Definition and Testing of Heat Pump Specifications
• U.S. EPA – What Is CHP?
• EUR-Lex – Directive (EU) 2023/1791 on energy efficiency