Interpretation of cost estimates in the process industry
How project maturity, risk and uncertainty shape CAPEX estimates
In process industries and project developments, investment cost estimates (CAPEX) are among one of the most important deliverables. CAPEX costs together with the operating costs (OPEX), have a significant impact on the business.
Cost estimate scope and time set the baseline for a project and are essential for planning, budgeting, earned value and the decision-making process throughout a project's lifecycle. These cost estimates vary in accuracy and detail depending on the maturity of the engineering and available information regarding deliverables during Front-End Loading stages (FEL 1-3) and into execution. The rationale behind investment cost estimates may appear straightforward: to ascertain the investment's cost, assess its profitability, timeline, scope, and strategies in the different phases and to strengthen the case for arranging financing with investors. This was also highlighted in our previous insight article, “Investment Cost Calculations in the Process Industry”. In this article we would like to further elaborate regarding the understanding and interpretation of the cost estimate.
Structure and basis of cost estimate
The cost in the CAPEX estimates are in AFRY categorized as
- Direct costs
- Indirect costs
- Risk (allowance)
- Contingency
Direct costs include equipment, materials, and labor associated with the physical construction of process facilities, infrastructure, buildings, etc. Indirect costs include costs associated with the implementation of the project and are incurred by the owner or consultant during engineering, procurement, construction, and commissioning of the project. Contingencies are included in the estimate as a separate line item, an amount added to the estimate to cover unforeseen costs or uncertainties within the project’s defined scope (see detailed description below).
CAPEX estimate should include two main structures, Work breakdown structure (WBS) and Cost breakdown structure (CBS). The WBS describes the deliverable-oriented hierarchical structure of the phases, disciplines and engineering work packages and is part of the planning baseline. The CBS describes the allocation of the costs in each of the WBS items to set the execution budget and should match the structure of the WBS that enables easy execution, earned value management, cost monitoring with a baseline and benchmarking.
Table 1 presents an example of a CAPEX overview for a project in the process industry. The presented structure is common in the chemical industry, but specific structures may vary between sectors such as mining/metals and pulp and paper, as well as between individual companies.
A clear understanding of the scope is essential for producing a reliable cost estimate. To compensate an unclear scope the cost estimator can calculate the risks and contingency. The more effort spent in structuring and fleshing out the scope (cost and time), the lower fluctuations should be expected in execution phase (cost, scope and time).
The estimate is derived from engineering work packages, supplier information and assumptions presented in the drawings, scope definition, reports, and WBS including all associated items defined in the scope of work. The total investment cost (TIC) estimates are then summarized.
Planning for the unknown: The role of contingency in cost estimates
Contingency is a vital part of the cost estimate. Contingency is defined as funds added to the cost estimate to compensate for estimate inaccuracies or unforeseen costs due to uncertainties in the project definition and/or engineering maturity level. Contingency does not cover scope changes, force majeure, or other project risks.
Contingency reflects estimating uncertainty, technology maturity, and vendor quote variability, not inefficiency or over design. But what exactly is it? What does it mean? And how is it decided?
Contingency is normally included in the estimate as a percentage of the total project cost to cover uncertainties and unknown risks. As the various Front-End Loading engineering stages present a different level of engineering maturity, it should be expected that the contingency will decrease with increased engineering maturity. Contingency can also be affected if the project is greenfield or brownfield, brownfield is normally associated with higher unknowns in the field. The contingency determination should always be tailored to each project phase and requirements.
Examples of what categories contingency could be:
- Errors and omissions
- Minor design development errors or incorrect quantities
- Unexpected vendor problems
- Procurement errors
- Any unexpected events
Contingency excludes:
- Request of scope changes such as changes in capacity and the “end product” specification, building size etc.
- Extraordinary events such as major strikes and natural disasters, “force majeure” and “act of god”
- Escalation and currency effects
It is essential to ensure that contingency funds are used for their intended purposes. It is important to keep in mind that the intention of the contingency is to compensate for the unknowns and not a lack of performance or progress. Contingency is not a substitute for proper cost estimating or project planning. It should not be added to the Base Estimate in budgets, cost plans or forecasts as an alternative to sound, robust, properly founded cost estimating.
How shall we handle allowance (known risks)? And what is it for?
Allowances are added on costs estimated by technical disciplines and/or cost estimator when applicable and required. An allowance is a cost element included in the base estimate to cover known cost items which need further engineering to allow for costing. They are expected to be used during the implementation of the project. Allowances will usually decrease as the project scope and design mature. Allowances are generally not specified as separate costs in the cost breakdown structure; instead, they are included within various engineering estimates.
Example of allowance are:
- Pumps on post 2900 from cost overview
- Allowance of pumps can be for an example: country of purchase which impacts tariffs, shipping conditions, guarantee etc.
- Piping material on post 4000 from cost overview
- Allowance for piping parts can be material that you calculate by factors that include corrosion and erosion allowance (to account for material loss over time), machining allowance (for grinding or finishing), and fitting allowance and effectivity at site.
- Allowance for quantity growth.
As you know, Contingency and Allowance are two different things, but both are very important for the total investment cost, see table 2 under.
AFRY has data from previous projects that details the contingency and risk levels across different project phases. Generally, projects have higher contingency levels due to lower project maturity, which matches the recommendations provided by AACE.
However, what we also notice is that the project complexity and technology level must be taken into consideration. Particular emphasis should be made to projects involving novel technologies, early-stage developments, and start up organizations, where AFRY reference projects may not be applicable and traditional benchmarking and Class based estimating frameworks offer limited support.
Risk management and contingency methods in industrial CAPEX projects
Continually evaluating risk and working on mitigations proactively will improve cost certainty.
By identifying risks and assessing their probability and calculating the potential financial impact using high and low figures, project teams can implement effective mitigation strategies to reduce both their likelihood and any impact, this should be carried out continuously per project phase and deliverable.
Contingency methods in industrial CAPEX projects
Contingency estimating is an integral component of the risk management process. The view on how to estimate the contingency varies and there are several different methods used for estimating contingency.
Expert Judgement: The judgement must have a strong basis in experience and be backed up by competency in risk management and analysis.
Predetermined guidelines: A set of guidelines where contingency levels are predetermined to match specific accuracy ranges and project complexities.
Simulation analysis (Monte Carlo): The cost model is usually a summary of estimated costs at some level of detail. Each cost element is assessed with a range and distribution that is assigned by the team based on their understanding of risks involved. The range estimation output is a total cost distribution to support decision making.
As illustrated in Figure 1 in Hollmann’s The Monte Carlo Challenge: A Better Approach1; by employing a probabilistic approach, contingency refers to the additional amount of money that needs to be added to the point estimate (the best estimate of all known items) to achieve a cost value that gives management a reasonable level of confidence (for example, P50). P50 stands for the 50th percentile in a probability distribution of possible outcomes. It means there is a 50% chance that the actual cost will be less than or equal to this value, and a 50% chance it will be higher.
One challenge in assessing risk-driving factors into contingency is the difficulty of gaining full insight and understanding of underlying issues, such as poor scope definition, and how these affect individual cost items. Therefore, AFRY are working with advanced work packages which improve scope definition, reducing the chance of missing or double counting items. The use of advanced work packages allows for more accurate cost estimating, clearer risk identification, all of which contributes to reducing uncertainty and improving contingency.
It is crucial to recognize and be aware of these underlying risk factors and to consider them when making an overall contingency assessment. When working with line-item costs and focusing on cost reduction or value improvement, risk factors also need to be considered. Value and risk management are closely related. Before starting contingency estimation, the risk drivers should be identified.
- Begin by identifying and understanding the key risk drivers.
- Differentiate between systemic risks and project-specific risks.
- Systemic risk affects an entire industry or large parts of it and is linked to broad economic, political, or social factors, such as interest rate fluctuations, currency exchange volatility, or geopolitical events. Project risk is specific to an individual project and may include issues such as delays in key equipment delivery or cost overruns.
- Use empirically-based stochastic models to address systemic risk drivers.
- Apply methods that explicitly link risk drivers to cost outcomes for project-specific risks.
- If using Monte Carlo simulations, ensure that dependencies between variables are properly accounted for. Monte Carlo analysis should preferably be used only when the project's maturity level is sufficiently high.
Risks are things that drive uncertainty of future outcomes and should not be confused with things that are simply higher in cost.
From cost estimate to total project budget
The calculated investment cost baseline has a certain level of accuracy, which indicates how much the estimated cost may deviate from the final actual cost of the completed project or activity. Rather than representing a single fixed number, a cost estimate reflects a range of possible outcomes. Estimate accuracy is derived from a probabilistic assessment of uncertainties and risks, forecasting the potential deviation of the final cost from the selected point estimate.
The estimate is derived from engineering material and assumptions, and the accuracy varies depending on the project maturity level. The AACE cost estimate classes and their expected accuracy range is summarized in Table 3. The accuracy range for a given estimate represents 80% confidence interval of actual costs from the cost estimate after application of appropriate contingency (typically to achieve a 50% probability of project cost overrun versus underrun) for a given scope.
A CAPEX estimate that is accurately calculated based on a defined scope represents the mean value within the statistical distribution of potential outcomes. As illustrated in Figure 1 in Hollmann’s The Monte Carlo Challenge: A Better Approach, each CAPEX estimate is associated with a confidence range determined by the expected level of accuracy. With a given probability, we can specify that the final CAPEX cost will fall within this predefined interval.
For project owners or portfolio managers, determining the appropriate project budget based on the CAPEX estimate raises important considerations. How much funding should be requested from project approvers, sponsors, or boards? For example, if the CAPEX estimate is 100 MSEK with a ±10% range at 80% confidence, the cost interval is 90–110 MSEK. However, there is still a 20% probability that the actual cost will fall outside this range. Allocating a project budget of 100 MSEK implies a 50% risk of overrun; organizations must assess whether such a risk level is acceptable.
To mitigate the risk of exceeding the budget, an “additional risk add-on” may be applied to the CAPEX estimate. The decision regarding the size of this add-on should be made by the project manager or director, taking into account the organization's risk management philosophy. Applying an additional risk add-on across a large project portfolio can result in unnecessary capital being tied up, which could otherwise be allocated elsewhere. This approach may also impact individual project profitability calculations, potentially leading to viable projects being rejected due to less favorable financial metrics.
In organizations managing extensive project portfolios, generally no additional risk add-ons are applied. Statistically, approximately half of the projects will exceed their budgets, while the other half will underrun, resulting in a balanced overall portfolio without excess funds held in reserve. On the other hand, if you have few or even a single project and your organizational policy is “no overruns allowed”, presenting CAPEX average, with no additional risk add-on, as total budget is risky.
In addition to the CAPEX estimate, the project owner should typically also account for costs such as financial expenses, taxes, land acquisition, completed feasibility and preliminary studies, start-up preparations and capital, as well as internal costs related to training operations and maintenance teams and systems. A schematic overview of the total project budget and its different parts is visualized in Figure 3.
Conclusion
Instead of indicating a precise figure, a cost estimate reflects the uncertainty inherent in forecasting and represents a spectrum of possible results. Estimating investment costs is a critical component of successful project planning and decision-making. However, it is a process that involves many connected factors and requires thoughtful coordination. To ensure accuracy and reliability, it is essential to approach cost estimation in a structured and methodical way. Identifying and understanding your unknowns is key to securing investments, as it enables informed estimation and calculation.
A comprehensive cost calculation must consider both direct and indirect expenses, including materials, labor, equipment, etc. Equally important is the assessment of project maturity. Projects in their early stages often involve higher levels of uncertainty, making it more difficult to produce accurate estimates. As a project progresses and more information becomes available, the precision of cost estimates improves. Therefore, it is essential to align the level of detail in the cost estimate with the project's stage of development.
Risk factors must also be carefully evaluated and incorporated into the cost estimating. By identifying and quantifying potential risks, the project organization can estimate appropriate contingency level and make more informed investment decisions. A cost estimate is based on several different assumptions and estimations, which provide valuable input into how the calculated costs should be interpreted. The estimate will be used not only within the project but also at various levels across the company. As outlined in the previous article, “Investment Cost Calculations in the Process Industry”, several other factors may also affect both the cost estimate itself and the methodology used to develop it. Therefore, it is crucial to ensure transparency regarding how the costs have been calculated, the assumptions made, and the prerequisites behind them.
In summary, investment cost estimation is not just about numbers, it is about understanding the full scope of a project, its context, and its potential challenges. A structured approach that incorporates transparency, project maturity and risk analysis is key to developing realistic investment cost estimates. A Basis of Estimate Report needs to be produced that provides clear and detailed explanation of how a project cost estimate was developed. This includes outlining the scope, data sources used, assumptions, risks, contingency etc. and the reasoning used to build the numbers. To produce a cost estimate an estimate validation exercise must also be carried out. The Estimate Review requires a qualitative assessment of the estimate’s conformance with technical requirements and industry best practices coupled with a quantitative assessment of the estimate’s competitiveness and alignment with business targets using historical and industry data.
This article is written by
Jonas Kihlman, Head of Development area New Technologies & Start-up Companies
Joseph Corrigan, Senior Cost Estimator
Jonas Sporrong, VP & Head of Pulp & Paper
