7 Practical Points for Estimating Utilization Loss in PVSyst

Table of Contents

• Why estimating utilization loss in PVSyst becomes important

• Practical Point 1: Don’t treat utilization loss as a “placeholder fixed value”

• Practical Point 2: Distinguish planned stops from unplanned stops

• Practical Point 3: Base assumptions on site conditions and maintenance arrangements

• Practical Point 4: Identify main causes of stops by equipment system

• Practical Point 5: Reflect monitoring setup and recovery time in utilization loss

• Practical Point 6: Don’t treat first year, steady years, and long-term operation the same

• Practical Point 7: Back-check with comparative simulations and result screens

• How to turn PVSyst utilization loss estimates into practical outcomes

Why estimating utilization loss in PVSyst becomes important

When running PV simulations in PVSyst, practitioners tend to focus first on items that directly affect generation: module capacity, azimuth, tilt, PCS conditions, shading, temperature, and wiring. On the other hand, utilization loss often looks like an auxiliary field where you enter a summarized number later, and it can be treated lightly. In reality, however, no matter how ideal the design assumptions, the impression of annual generation can change greatly depending on how you account for the time equipment is stopped or not fully operating. In other words, utilization loss is not a mere add-on at the end of generation calculations but an important assumption that brings the system back to reality.

Here, utilization loss refers to losses that occur even when conditions are otherwise favorable for generation, due to things like inspections, failures, replacements, communication problems, troubles around transformer/substation equipment, or delays in restarting operations—i.e., times when the system cannot operate fully. This is different in nature from losses caused by natural conditions such as low irradiance or shading. Because utilization loss is influenced not only by system composition but also by operation schemes, maintenance arrangements, site accessibility, and ease of recovery, it is a practical issue that spans both design and operation.

This item becomes important in practice because it can affect the relative merits of comparison options. For example, a design that shows slightly higher theoretical first-year generation may be disadvantaged in expected annual actual generation if inspection routes are poor and recovery from stops tends to take a long time. Conversely, a layout that is orderly and easy to maintain may appear modest in capacity but, when utilization loss is taken into account, can look robust and preferable. If you use PVSyst in practice, closing the gap between theoretical desk-based generation and actual operability is critical.

Also, how you handle utilization loss affects persuasiveness in internal briefings and proposal materials. Showing a large generation number is easy, but if you cannot explain the operational assumptions underlying that number, it becomes a weak basis for decision-making. Numbers that include utilization loss and are well organized are more likely to be evaluated as realistic and easy to explain, even if the generation looks somewhat conservative. Estimating utilization loss in PVSyst is not about making numbers conservative for their own sake but about tying design and operation into a single plan.

Practical Point 1: Don’t treat utilization loss as a “placeholder fixed value”

When estimating utilization loss, the first thing to avoid is inserting a “safe-looking fixed value” and calling it done. In practice, people often want to move forward quickly to get an overall view of generation and therefore apply a fixed percentage. That approach can be convenient in early stages, but if you proceed to comparisons and explanations without scrutinizing that number for the specific project, the meaning of the results becomes thin. What really matters in PVSyst is not merely that a number is entered but whether that number is a natural assumption for the project.

For example, it’s not necessarily appropriate to apply the same utilization loss to a project with a cramped site and limited maintenance routes as to one with wide aisles and well-organized equipment zones. Similarly, response speed after stops will differ between a mountain site with long access times and a flat site easy to inspect. If you apply the same fixed value across all projects, differences in maintainability won’t be reflected in generation comparisons. Because PVSyst is a tool for comparing project-specific differences, an overly simplistic fixed value weakens that comparative power.

Also, using a fixed value obscures where improvement is possible. If the causes of large utilization loss aren’t sorted into whether they’re inspection interruptions, unplanned stops, or recovery delays, it’s hard to tell what to change to stabilize annual generation. In practice, more important than a few percentage points of loss is knowing which operational conditions can be modified to reduce that loss. If you enter utilization loss in PVSyst, you should be aware of what that number represents rather than relying on a simple fixed value.

As a countermeasure, roughly organize what kinds of stoppages and capacity reductions are likely on the project before entering utilization loss. Even if you can’t produce a detailed breakdown, dividing it into categories such as inspection-related, failure-related, and recovery-delay-related will give the number meaning. To handle utilization loss in PVSyst without hesitation, you should be able to explain what the single fixed number represents before applying it.

Practical Point 2: Distinguish planned stops from unplanned stops

When estimating utilization loss in practice, it’s important to separate planned stops from unplanned stops. Both show up as reduced availability, but their implications for design and countermeasures differ. Planned stops include inspections, cleaning, and scheduled maintenance that can be anticipated to some extent. Unplanned stops include failures, communication anomalies, and delayed recovery actions, whose timing and duration are hard to predict. If you lump these together into one utilization loss figure, you may be able to place a number but lose sight of where to focus operational improvements.

For instance, if a plan is dominated by planned stops, losses might be reducible by organizing maintenance schedules and work procedures. If an option has high unplanned stop risk, you need to consider equipment clarity, accessibility, monitoring setup, and spare parts management. Even if PVSyst ultimately aggregates both into a single loss, practitioners should understand that the backgrounds differ. What matters is not just the magnitude of the number but which type of stoppage it reflects.

This separation also helps when reading comparison options. One design might have narrow aisles making routine inspections inefficient, while another might have fragmented zones that complicate isolation and recovery during unplanned stops. The same one-percent difference can mean different things in practice. If you compare utilization loss in PVSyst, consider not only the total magnitude but also which type of stoppage is the main cause—this will make design decisions far more concrete.

As a measure, when setting utilization loss, organize roughly how much of the number you expect to come from planned stops versus unplanned stops. With that in mind, when you review comparison options you can judge which design is easier to inspect and which is easier to recover. Distinguishing types of stops is essential for practical estimation of utilization loss in PVSyst.

Practical Point 3: Base assumptions on site conditions and maintenance arrangements

To reasonably estimate utilization loss, you must base it on site conditions and maintenance arrangements. In practice, people may think similar-sized systems should have similar utilization loss, but if site access or maintainability differ, the time from stop to recovery can vary greatly. PVSyst is a simulation tool, but numbers used in practice cannot ignore on-site operational conditions.

For example, a project close to a road with easy access for service vehicles and straightforward equipment zoning with sufficient aisles will be relatively easy to handle for both routine inspections and abnormal responses. Conversely, an array divided into multiple areas with slope or aisle constraints and a layout that’s hard to grasp may take longer to recover from the same fault. Utilization loss should capture these operational differences rather than being seen as a simple equipment performance gap.

Maintenance arrangements also matter. How routine inspections are scheduled, how quickly remote monitoring picks up anomalies, and how much time you assume to reach the site all affect utilization loss even for the same equipment. While PVSyst doesn’t let you enter all these factors directly in numeric detail, you should organize them as the background to the utilization loss figure. Doing so will give comparative differences real on-site relevance.

As a countermeasure, before setting utilization loss, describe in words how maintainable the project is: Is access easy? Are zones well organized? Are aisles sufficient? Is inspection straightforward? Is remote monitoring likely to detect issues quickly? Even considering these aspects changes your sense of the appropriate number. To handle utilization loss correctly in PVSyst, you must view the system not only by theoretical performance but as a system that will be maintained on site.

Practical Point 4: Identify main causes of stops by equipment system

When estimating utilization loss, it’s also important to identify stoppage causes by equipment system. In practice, people sometimes treat equipment stoppages as a vague single risk, but module, PCS, transformer/substation equipment, communications, monitoring, and on-site work issues all differ in how they stop and how difficult they are to recover. Even if PVSyst aggregates them into a single utilization loss, separating causes before estimating makes the numerical basis much clearer.

For example, stops involving PCS or transformer/substation equipment can affect large portions of generation and have wide impact areas at recovery. On-site communication problems or monitoring anomalies might affect utilization more through delayed detection and recovery than by directly reducing generation. Module or string-level issues are influenced by how zones are partitioned and maintenance routes are organized. In short, utilization loss should be considered as the sum of how easily multiple systems stop and how easy they are to repair, not as a single equipment fault.

Organizing this also makes it easier to see strengths and weaknesses in comparison options. One option may have a straightforward PCS layout that is easy to isolate, while another may have many sections and be strong for individual responses but complicated in monitoring. Looking only at annual generation differences in PVSyst can miss these operational differences, but listing stoppage causes by system will give utilization loss numbers meaning.

As a measure, before setting utilization loss, briefly list the main stoppage causes by equipment system and consider where the bottleneck is likely to be for the project. You don’t need to perform strict probability assessments—just organizing the types of stoppages will make PVSyst figures much more practice-oriented. Even if you ultimately use a single aggregate coefficient, it’s important to understand the stoppage causes behind it.

Practical Point 5: Reflect monitoring setup and recovery time in utilization loss

To estimate utilization loss realistically, you must incorporate the monitoring setup and recovery time. In practice, attention tends to focus on fault occurrence itself, but what really affects annual generation is how quickly you detect anomalies and how quickly you recover. PVSyst is not a tool to evaluate monitoring systems in detail, but it leaves the decision of which operational assumptions to reflect in utilization loss to the user. For that reason, failing to consider differences in monitoring leads numbers to lean too much on desk-based assumptions.

For example, a project with well-developed remote monitoring and a short flow from anomaly detection to on-site response will have different annual impacts from faults than a project that relies on inspections and whose anomaly detection is slower. Recovery time also varies depending on whether parts can be quickly sourced, how long site access takes, and whether you can isolate stopped sections easily. In practice, the key is not the fault probability itself but the sense of time from stopping to resuming. When estimating utilization loss in PVSyst, the number should reflect these operational differences.

Moreover, monitoring and recovery assumptions influence the meaning of design comparisons. One option may have well-organized zones that make isolation easy; another may be highly aggregated so a single stop affects a wide area. Ignoring these differences and applying the same utilization loss conceals practical differences in operability from generation comparisons. If you want to connect PVSyst numbers to design decisions, you must incorporate not only how stops occur but how easily they can be restored.

As a countermeasure, when setting utilization loss, at least roughly organize the project’s monitoring and recovery setup: Is anomaly detection fast? Does reaching the site take long? Are part replacements straightforward? Just considering these perspectives significantly changes the sense of the number to enter. When estimating utilization loss in PVSyst, think not only about components breaking but also about how you will return them to service.

Practical Point 6: Don’t treat first year, steady years, and long-term operation the same

When setting utilization loss, it’s important not to use the same assumptions for the first year, steady years, and long-term operation. In practice, people tend to produce one year of generation and treat that as a representative value. However, the nature of stoppages can change between the initial commissioning period and stable operation, and long-term maintenance and replacement assumptions may evolve. If you want practical numbers from PVSyst, be aware of these differences across time.

For example, the first year may include more adjustments and checks, so different stoppage causes may be prominent compared to steady years. In steady years, operations may stabilize and the balance between planned and unplanned stops can change. Over the long term, maintenance and replacement plans may also shift. Expressing all of these with the same utilization loss erases time-related differences and weakens the meaning of comparisons. When working with utilization loss in PVSyst, you should at least clarify which period the number represents.

Thinking in time slices also deepens how you view comparison options. Whether the comparison emphasizes the first year, steady-year stability, or long-term assurance will change the evaluation of the same design. In practice, if it’s unclear which time axis the figures represent, the meaning of comparisons can wobble later. If you want PVSyst results to be explainable, organize the relationship between the time axis and utilization loss.

As a countermeasure, before entering utilization loss, be clear whether the number represents the first year, a steady year, or assumptions closer to long-term operation. If needed, separate which time axis you emphasize for each comparison option. To estimate utilization loss practically in PVSyst, do not treat one year’s number as a universal representative.

Practical Point 7: Back-check with comparative simulations and result screens

Finally, it is important to back-check the plausibility of utilization loss using comparative simulations and result screens. In PVSyst, entering utilization loss yields annual generation, and a tidy number can tempt you to accept it. But in practice you must question whether that utilization loss is truly a natural assumption by looking at the results, because stoppage and recovery assumptions are hard to see and numbers can easily be over- or under-estimated.

For example, if two designs that shouldn’t differ much based on equipment configuration or array layout show a large annual generation gap, utilization loss assumptions may be exerting too much influence. Conversely, if designs with clear differences in maintainability or monitoring show almost no difference in results, you may be smoothing too much. The point of comparative PVSyst analyses is not only to confirm generation differences but to judge whether those differences are natural in light of the underlying assumptions.

Getting into the habit of back-checking via result screens also strengthens internal explanations. You will be able to explain why you set a given utilization loss, why a design appears preferable in steady years, and why impressions change between first year and long term—essentially explaining the numbers’ background. In practice, being able to explain numbers can matter more than strictly proving their correctness. PVSyst is both an input tool and a tool for validating the plausibility of assumptions.

As a measure, after setting utilization loss, always check whether the generation differences between comparison options, the differences in stoppage conditions, and the differences in maintenance setups align so that the numbers look natural. If something feels off, revisit not only equipment conditions but also the assumptions behind utilization loss. Correctly estimating utilization loss in PVSyst requires not simply entering a value but repeatedly questioning its plausibility based on results—that iterative check is the last practical differentiator.

How to turn PVSyst utilization loss estimates into practical outcomes

The common thread among the seven practical points above is not to let utilization loss be a mere correction factor. Don’t use a fixed value; separate planned and unplanned stops; base assumptions on site conditions and maintenance arrangements; organize stoppage causes by equipment system; reflect monitoring and recovery times; treat first-year, steady-year, and long-term operation differently; and finally back-check with comparative simulations and result screens. When you follow this flow, utilization loss in PVSyst becomes not just a number that slightly adjusts annual generation but an assumption that expresses the system’s operational realism.

What really matters to practitioners is not selecting the option that outputs the highest annual generation. What adds value is being able to explain why you chose that utilization loss for the project. If you have organized aspects such as ease of inspection, ease of recovery, zone coherency, and monitoring differences, simulation results become more than desk-based numbers—they become design proposals that incorporate operation. Conversely, careless utilization loss assumptions can make even attractive numbers fall apart in later validation.

Moreover, improving utilization loss accuracy requires not ending the process at desk simulations. If site boundary, aisle conditions, slopes, buildings, string divisions, and monitoring methods are ambiguous, assumptions about stoppage and recovery will be ambiguous as well. To connect PVSyst results to practice, you need to iterate between site understanding and simulation to confirm what degree of availability reduction is natural. Utilization loss is both a coefficient on the screen and a reflection of how easy the system is to operate on site.

In that sense, when you want to more reliably confirm positions and coordinates on site, using iPhone-mounted GNSS high-precision positioning devices such as LRTK is an effective idea. Better-organized on-site positional information and site conditions make it easier to set clearer assumptions for array layout, maintenance aisles, and equipment zones when estimating utilization loss in PVSyst. If you can raise desk-comparison accuracy with PVSyst and support on-site understanding with LRTK, setting utilization loss becomes less of a mere loss input and more of an on-site–rooted practical judgment. Carefully estimating utilization loss not only improves the accuracy of generation forecasts but also enhances the design and operational decision-making that links desk work to the field.