Energy security for a business does not mean that a facility must be self-sufficient and operate entirely off-grid. In practice, it is about whether the company can keep its most important processes running at an acceptable level, shut them down safely and resume operations efficiently when the energy supply is interrupted or restricted, or when its quality parameters deteriorate.
An energy security assessment therefore should not begin with the question: “do we need a generator, energy storage or photovoltaics?” The first step is to determine which processes are critical, how long they can remain without power, which energy carriers and equipment they depend on, and what the consequences of stopping them would be. Only then can the risk be assessed and a decision made on whether additional safeguards are needed.
What does energy security mean for a business?
From the perspective of a specific company, energy security can be understood as the ability to maintain continuity of critical operations during disruptions to energy supply and to limit the consequences of such disruptions to a level accepted by the organisation.
In practice, four questions need to be answered:
- What must keep operating? Which processes, installations and loads are critical to production, personnel safety, product quality, the environment, IT or customer service?
- How long can it be unavailable? Is the acceptable interruption a fraction of a second, a few minutes, an hour or an entire shift?
- What can interrupt operation? A grid failure, transformer failure, switchgear failure, control system failure, loss of gas supply, lack of generator fuel, a connection-capacity constraint, or failure of an on-site generation source?
- What safeguards are already in place? A second supply, UPS, generator, energy storage, equipment redundancy, load-shedding capability, emergency procedure, fuel reserve or service contracts?
It is also worth distinguishing energy security from energy independence. Full independence is one possible strategy, but it is not a prerequisite for a high level of security. A facility can be well protected while relying primarily on the grid if it has adequate redundancy, supplies critical loads appropriately and has a prepared response for outages. Conversely, having an on-site energy source does not guarantee security if that source itself has a single point of failure or cannot operate in island mode.
Process first, load second: how to identify what is truly critical
The largest energy load is not always the most critical one. A 300 kW motor may be shut down without major consequences, while a PLC, server, cooling pump or ventilation system rated at only a few kilowatts may determine whether the entire installation can be shut down safely.
The analysis should therefore proceed from the process to energy, not the other way around. For each important process, record at least:
| Assessment element | Practical question |
|---|---|
| Process or function | What must the company keep operating or shut down safely? |
| Energy carrier | Electricity, gas, heat, steam, fuel or another auxiliary medium? |
| Supply point | Which transformer, switchgear, pipeline, source or system supplies the process? |
| Critical loads | Which equipment is essential to maintain operation or achieve a controlled shutdown? |
| Acceptable interruption time | How long can the interruption last before significant losses or risks arise? |
| Restart conditions | Is restart immediate, or does it require heating, cooling, venting, cleaning or a new quality check? |
| Impact of energy loss | Downtime, missed shipment, loss of a product batch, occupational safety hazard, emissions, equipment damage or data loss? |
A well-prepared dependency map often shows that a critical process does not depend on a single device but on an entire chain. A production line may have an electrical supply, but without compressed air, process cooling, ventilation or controls it will still be unable to operate.
Three levels of load criticality
For an initial screening, loads can be divided into three groups:
- uninterruptible or highly sensitive – controls, IT, automation, protection systems and selected measurement systems; even a very short interruption or voltage dip can stop them;
- requiring short-term backup – equipment needed for a controlled process shutdown or to maintain temperature, pressure, ventilation, cooling or safety;
- temporarily interruptible – loads whose shutdown does not immediately cause significant losses and which can be restarted when power returns.
This classification is far more useful than trying to back up an entire facility simply because its peak demand is a certain number of megawatts.
The consequence of a failure is not just the energy that was not consumed
The cost of an energy supply interruption rarely ends with the value of a few unused megawatt-hours. For a business, the chain of consequences is more important.
An interruption may cause, among other things:
- production or service stoppage;
- loss of raw material, work in progress or a batch requiring continuous process conditions;
- additional restart time after the energy supply returns;
- cleaning, reheating, cooling or recalibration;
- loss of product quality parameters;
- employee downtime and additional labour costs;
- failure to complete a shipment or meet customer commitments;
- risks to people, the environment or infrastructure;
- data loss or interruption of IT and automation systems.
A two-hour power outage therefore does not necessarily mean two hours of downtime. If returning the process to stable production takes another three hours, the consequence is five hours of lost production capacity, not just the duration of the grid outage itself.
What do Polish regulations say about interruptions in electricity supply?
The Polish Regulation on the detailed conditions for the operation of the electricity system distinguishes between planned and unplanned interruptions and, by duration, between momentary interruptions (micro-outages) of up to 1 second, short interruptions of more than 1 second and up to 3 minutes, long interruptions of more than 3 minutes and up to 12 hours, very long interruptions of more than 12 and up to 24 hours, and catastrophic interruptions lasting more than 24 hours. For connection groups I–III and VI, the permissible interruption durations are specified in the electricity transmission or distribution service agreement or in a comprehensive agreement. For groups IV and V, a single planned interruption may not exceed 16 hours and an unplanned interruption may not exceed 24 hours; the total annual duration of long and very long interruptions may not exceed 35 hours for planned interruptions and 48 hours for unplanned interruptions, respectively. The current consolidated text of the regulation in ELI should be read together with the 2026 amendment.
From a company’s perspective, the key conclusion is different: the interruption time permitted by Polish regulations or by a contract is not the same as the interruption time acceptable to a technological process. A business may suffer significant consequences after only seconds or minutes, even if the event still falls within the categories permitted for grid operation.
Seven risk areas worth checking
1. External supply and power quality
The first area is how energy reaches the facility. What matters is not only the number of connections, but also their actual independence, shared grid components, voltage level, outage history and process sensitivity to voltage dips, micro-outages and deviations in power-quality parameters.
Two cables do not provide full redundancy if both run through the same substation, switchgear section or another common component whose failure disconnects the entire facility.
2. Internal energy infrastructure
A failure does not have to occur on the grid operator’s side. A single transformer, main switchboard, busbar, protection device, cable, contactor, automatic transfer switch (ATS) or automation component can form a single point of failure shared by several critical processes.
The assessment should therefore also cover technical condition, protection selectivity, availability of spare parts and service support, the possibility of bypassing a failed component, and the time needed to restore supply after an internal failure.
3. Dependence on a single energy carrier or supplier
The risk is not limited to electricity. A facility may be highly dependent on gas, fuel, district heating or a specific process medium. It is worth checking whether the loss of one carrier stops all production and whether there is a technically feasible alternative operating mode.
Diversification, however, does not automatically mean building a second source. If an event is unlikely, its consequences are limited and the process can be shut down safely, additional infrastructure may not be justified.
4. Connection constraints and lack of headroom for growth
Energy security also has a forward-looking dimension. If a company plans process electrification, a new production line, expanded refrigeration capacity, heat pumps, fleet charging or increased production, it should verify whether the existing grid connection and internal infrastructure can support the new demand profile.
In Poland, connection conditions specify, among other things, the connection capacity and the scope of network modifications required for the connection. The provision concerning an application for connection conditions also applies accordingly when an entity being connected or already connected increases its required connection capacity or changes the existing technical conditions and operating parameters of its equipment, installations or network. The Polish Energy Regulatory Office (URE) notes that the obligation to conclude a connection agreement depends on the existence of technical and economic conditions for connection and energy delivery and on the applicant meeting the connection requirements. URE explains the procedure in the event of a refusal to connect.
The scale of these constraints in Poland is not merely theoretical. In its report for 2023–2024, URE recorded 15,277 refusals to conclude electricity-grid connection agreements, relating to more than 157 GW in total. The regulator also notes that the total capacity may be overstated because applications for the same projects can be submitted multiple times. Nevertheless, URE data show why connection feasibility should be verified sufficiently early in projects that increase power demand.
5. On-site generation and energy storage – resilience depends on operating mode
Photovoltaics, cogeneration, biogas, a wind turbine, energy storage or another source can reduce dependence on the grid, but not every installation can supply power during an outage.
Many photovoltaic systems are designed for grid-parallel operation and disconnect when grid voltage is lost for safety reasons. Island operation requires suitable architecture, automation and a source capable of maintaining voltage and balancing power. The U.S. Department of Energy notes that PV systems with storage can transition to island mode if they are designed to do so. DOE: Solar and Resilience Basics.
Similarly, cogeneration can improve resilience, but only if fuel is available, the system can start appropriately and the architecture allows it to supply critical loads. Energy storage, in turn, requires suitable power, usable capacity, state of charge and control logic.
6. Controls, automation and auxiliary loads
Energy may be physically available, yet the process still may not start if controls, communications, BMS/EMS, PLCs, safety systems or auxiliary equipment are unavailable. The analysis must therefore consider not only the main loads but also the conditions required for their operation.
This is one reason why energy monitoring and management can support energy security: they provide better information about loads, installation status and load behaviour, although the monitoring system itself does not replace a backup power source.
7. Organisation of the response during an outage
The final area is often the least expensive to improve, while at the same time determining whether the available safeguards are used correctly. The company should know:
- who decides to start backup power;
- which loads have priority and in what sequence they are energised;
- how the process is shut down safely and restarted;
- who contacts the grid operator, service provider and fuel supplier;
- whether fuel, parts and personnel are actually available during an emergency;
- when the UPS, generator, ATS or transfer procedure was last tested;
- where failures and their actual consequences are recorded.
In its guidance for businesses, the DOE highlights, among other things, deciding when to activate backup power and confirming that fuel and a supplier are available during an emergency. DOE: Business Owners – Respond to an Energy Emergency.
How to build a simple risk matrix: probability × consequence
An initial assessment does not require a complex probabilistic model. A consistent five-point scale for probability and consequence is sufficient. This is a working approach: thresholds should be adapted to the specific organisation and should not be treated as a legal or normative requirement.
Step 1. Assess the probability of the event
| Rating | Probability | How to interpret it |
|---|---|---|
| 1 | very low | no known events, good technical condition, low exposure |
| 2 | low | the event is possible but rare; effective safeguards are in place |
| 3 | medium | the event is credible; it has occurred locally or there are significant technical indications |
| 4 | high | repeated disturbances, clear vulnerability or weak safeguards |
| 5 | very high | the event is frequent or expected under current operating conditions |
The assessment should be based on data: failure history, power-quality measurements, equipment condition, information from the operator, inspections, service availability and actual technical dependencies.
Step 2. Assess the consequences for the business
| Rating | Consequence | Example interpretation |
|---|---|---|
| 1 | negligible | local disturbance with no impact on critical operations |
| 2 | minor | short downtime in a single area, easy restart |
| 3 | moderate | partial restriction of production or services, measurable losses and recovery time |
| 4 | major | shutdown of an important process, product loss, serious delays or significant operational risk |
| 5 | critical | prolonged shutdown of critical operations, danger to people/the environment, significant losses or loss of the ability to meet obligations |
Step 3. Multiply the two values
Risk = Probability × Consequence
Example working interpretation of the score:
- 1–4 – low risk;
- 5–9 – moderate risk;
- 10–16 – high risk;
- 17–25 – very high risk.
The aim is not to calculate a “true” number for every risk. Above all, the matrix is intended to provide a consistent framework for comparing risks and show which scenarios require analysis first.
Auditor’s comment: before selecting a backup technology, two parameters must be known: the power required by critical loads and the period for which they must be maintained. Without this information, it is easy to design a safeguard that is too small, too large or protects the wrong loads—the ones that do not actually determine operational continuity.
Example: the facility draws 900 kW, but only 180 kW is critical
Assume a hypothetical facility whose demand during normal operation reaches 900 kW. Process analysis shows, however, that the entire production process does not have to be maintained during an outage. It is sufficient to support 180 kW of loads: controls, IT, selected pumps, cooling, ventilation and equipment required for a safe shutdown.
If the required backup time is 2 hours, the first approximation of the energy requirement is:
180 kW × 2 h = 360 kWh
This is not yet the storage capacity that should be purchased. Actual sizing must take into account, among other things, efficiency, the permissible state-of-charge range, reserve, ageing, temperature, inverter power, starting currents and possible load variability. It may also turn out that a generator, a hybrid system or backing up only part of the 180 kW is a better solution.
For the same facility, the risk register could look as follows:
| Scenario | P | C | Score | Conclusion |
|---|---|---|---|---|
| Unplanned grid power outage | 3 | 5 | 15 | high risk – safeguards and the required backup duration need to be checked |
| Failure of the only transformer supplying the critical switchboard | 2 | 5 | 10 | high risk – low P does not offset a critical consequence |
| Short voltage dip that stops the control system | 4 | 3 | 12 | high risk – a solution with a very fast response time may be required |
| Interruption of gas supply to a thermal process | 2 | 4 | 8 | moderate risk – the decision depends on the acceptable downtime and the availability of an alternative |
The values in this example are illustrative only. In a real facility, the score should be based on event history, process characteristics, technical data and the company’s adopted consequence criteria.
Backup power, storage, on-site generation or redundancy? Match the solution to the risk first
There is no single universal set of measures that every company should implement. What makes sense depends on the type of threat and the response time required by the process.
| Solution | What it can mitigate | Key limitation to verify |
|---|---|---|
| UPS | micro-outages, voltage dips and loss of power to controls/IT | short backup duration and limited power |
| Generator | longer interruption of supply to critical loads | start-up time, fuel, service, testing and transfer logic |
| Energy storage | fast backup, operation of critical loads, power support | power, usable energy, state of charge, island-mode capability |
| PV or wind | reduced grid import and partial diversification | variable generation; the installation itself may not operate during a grid outage |
| Cogeneration / dispatchable source | on-site energy generation and the ability to maintain critical functions | fuel availability, black-start capability, island operation, heat and power balance |
| Second connection / redundant transformer / sectionalising | reduced impact of a single failure | genuine independence of supply paths and appropriate automation are essential |
| Load control / load shedding | reduction in required backup capacity | correct priority hierarchy and safe shutdown sequence |
| Emergency plan and tests | faster response and fewer organisational errors | the procedure must be current, practised and assigned to specific people |
On-site sources can form part of an energy-system upgrade, but their value for security should be assessed separately from their economic viability as energy projects. Energy storage can provide backup, and cogeneration can increase the availability of a local source—provided the architecture is appropriate and fuel is available.
Other energy technologies should be assessed in the same way. Heat recovery can reduce demand for an external energy carrier and indirectly reduce process vulnerability, but it is not itself an emergency source. Photovoltaics and wind power increase on-site generation but depend on weather conditions. Biogas or hydrogen can serve as an energy carrier or fuel in specific systems, but using them to improve security only makes sense if the full chain of supply, storage, start-up and emergency operation has actually been assessed.
Energy efficiency also affects resilience – but it does not replace backup
Reducing energy demand can lower backup power requirements. If critical loads after optimisation require 150 kW instead of 200 kW, the required backup-source capacity may be lower. The DOE notes that improving facility efficiency can reduce the capacity required from backup systems and make it easier to maintain critical functions. FEMP: Resilience Planning and Valuation.
The two objectives should not be conflated, however. An energy-efficiency upgrade reduces consumption or demand but does not by itself eliminate the risk of a supply interruption. Likewise, storage installed solely for energy arbitrage or peak-demand reduction does not automatically provide backup—the control strategy and energy reserve must account for an outage scenario.
What should the output of an energy security assessment look like?
A good assessment does not have to end in a report hundreds of pages long. For many businesses, useful decision-making material can be built around a few specific summaries:
- Critical process map – what must operate, what can stop and what is needed for a safe shutdown.
- List of critical loads – power, operating profile, maximum acceptable interruption time, start-up conditions.
- Energy dependency map – connections, transformers, switchboards, sources, energy carriers, controls and common points of failure.
- Scenario register – external failure, internal failure, loss of an energy carrier, insufficient capacity, source failure, control error, organisational issue.
- P × C matrix – risk before safeguards and residual risk after existing measures are taken into account.
- Protection gaps – not as a shopping list, but as points where the current risk level exceeds the threshold accepted by the company.
- Priorities for further analysis – e.g. power-quality measurements, protection selectivity analysis, automatic transfer switch (ATS) test, generator test, storage analysis, increase in connection capacity, island-operation scenario.
When can the level of energy security be considered sufficient?
There is no single number suitable for every business. A chemical plant, cold store, data centre, logistics warehouse and office face entirely different consequences from the same one-hour interruption.
The level can be considered acceptable when the company has consciously defined:
- which functions are critical;
- the maximum acceptable duration of their unavailability;
- which scenarios could exceed that duration;
- the operational, financial and safety consequences;
- the safeguards already in place and their actual effectiveness;
- the residual risk after safeguards are applied;
- the threshold above which further action is required.
At some facilities, the outcome will be an expansion of backup power. At others, a UPS for controls, a dedicated switchboard for critical loads, an improved procedure, a second source of compressed air or an earlier application to increase connection capacity may be sufficient. Sometimes the right decision is also to accept the risk if the cost of reducing it is disproportionate to the consequences.
Ten questions for a quick self-assessment
If a company cannot answer several of the following questions, it is worth starting by organising the data:
- For which three processes would a loss of energy have the greatest consequences?
- What is the maximum acceptable interruption time for each of them?
- What is the total power of the loads that genuinely need to be maintained?
- Do we know the common single points of failure in the internal power supply?
- Can the on-site source or PV system operate during a grid outage, or only in parallel with the grid?
- Have the generator, UPS, ATS or energy storage system been tested under conditions close to an actual outage?
- Are fuel, service support and spare parts also available outside normal working hours?
- Will planned investments fit within the existing connection capacity and the capabilities of transformers and switchboards?
- Is it known which loads should be disconnected first when available power is limited?
- Can the company estimate the full impact of a one-hour interruption—including restart, quality losses and delays?
Frequently asked questions
Does a photovoltaic system improve a company’s energy security?
It can, but not automatically. A system designed solely for grid-parallel operation may shut down when grid voltage is lost. Supplying loads during an outage requires suitable island-mode architecture, controls and often storage or another source that stabilises the system.
Should every company have a generator?
No. A generator makes sense when it addresses a material risk and the required autonomy time. For some loads, a UPS or storage system will be better, while in businesses where downtime has limited consequences the cost of maintaining a generator may not be justified.
Does high reliability of the operator’s grid mean the risk is low?
Not always. Average network reliability indicators do not show how sensitive a specific process is to micro-outages, voltage dips or internal failures. Security must be assessed at the level of the specific facility and its technology.
How often should the assessment be updated?
There is no single universal interval for all companies. The assessment should be updated after a significant change in the process, power demand, connection, on-site source or control system, after a serious event, and whenever the level of risk accepted by the company changes. Independently of this, the failure register and safeguard tests should provide data for ongoing assessment.
Summary
Energy security for a business is not a list of technologies. It is a way of assessing whether a facility can maintain critical functions or manage an outage safely and then return to operation within an acceptable time.
The most practical sequence is: critical processes → acceptable interruption time → loads and energy carriers → failure scenarios → probability and consequences → existing safeguards → residual risk → decision on further action.
Starting with this map makes it easier for a company to avoid both underestimating risk and investing in expensive safeguards that do not protect the processes that are genuinely critical.
Sources and legal basis
- ELI – consolidated text of the Polish regulation on the detailed conditions for the operation of the electricity system, accessed 13 August 2026.
- ELI – amendment to the Polish regulation of 15 May 2026, accessed 13 August 2026.
- Polish Energy Regulatory Office (URE) – report on the operation of Poland’s electricity and gas sectors in 2023–2024, accessed 13 August 2026.
- Polish Energy Regulatory Office (URE) – refusal of connection to the electricity grid, accessed 13 August 2026.
- U.S. Department of Energy – Resilience Planning and Valuation, accessed 13 August 2026.
- U.S. Department of Energy – Solar and Resilience Basics, accessed 13 August 2026.
- U.S. Department of Energy – Business Owners: Respond to an Energy Emergency, accessed 13 August 2026.
