How to Assess the Value of Digital Enterprise Maintenance Investments

The fourth industrial revolution unveils a host of opportunities for improving your industrial maintenance activities, including the possibility to deploy and fully leverage predictive maintenance (PdM). The deployment of PdM is likely to require significant investments, including investments in hardware (e.g., sensors networks for automated data collection) and software (e.g., deep learning packages and asset management).

It may also become costly to procure consultants on BigData and Industrial IoT technologies. As a result, justification for these investments is at the forefront, as a means of convincing senior management to include these investments at the top of the corporate project portfolio.

Industrial organizations are looking into ways to calculate the Return On Investment (ROI) of any maintenance project. To determine the ROI, there are a number of calculator tools. ROI calculators provide a sound basis for understanding and quantifying the merits of an enterprise maintenance project, yet managers and maintenance engineers should be aware of associated limitations.

Understanding Capital Budgeting

Chief Financial Officers (CFOs) and financial managers are usually keen on using capital budgeting methods to assess the financial benefits and the overall profitability of investments. Most of the capital budgeting methods require a thorough consideration of the costs and the benefits of an investment.

In practice, the costs are known (or can be estimated) with high accuracy, while the benefits are 'approximated' in. This can result in uncertainty about the credibility of the outcomes.

As far as the costs are concerned, it is important to make sure that all the expenses are listed, including infrastructure, software, energy, labor and downtime costs. This means that the Total Cost of Ownership (TCO) of a new system or investment should be taken into account.

This can be quite challenging when costs are either overlooked or assumed.

Likewise, all projected benefits should be considered as well, including cost savings and revenue streams. Although benefits estimations may not be very accurate, they should be based on sound assumptions for operating a new system and its monetary performance.

An important element of any capital budgeting calculation is the time frame during which costs and benefits are taken into account.

For IT systems like asset management and enterprise systems, a period of five years is considered reasonable, given that this is a commonly used depreciation period for IT equipment and services.

However, the ROI of infrastructure investments is likely to be computed over rougher time scale estimates. Regardless of time frame, both capital budgeting and ROI calculation should take into account the “time value for money”—the fact that one dollar spent today will not equal a dollar spent three or four years later, due to inflation. As a result, when computing capital budgeting metrics, it’s always important to know if the estimates account for the time value of money.

With a detailed list of costs and benefits, various capital budgeting metrics can be calculated. In principle, these calculations require a good understanding of finance. However, most of the indicators are available as built-in functions in popular spreadsheet programs.

It’s important to note the following key metrics:

• The Payback Period, which indicates the time needed to yield benefits equal to the amount put in the investment. For example, with a $300,000 investment in a predictive maintenance system, a payback period of two years means that within two years, benefits of $300,000 will be produced, based on reduced downtimes and better Overall Equipment Efficiency (OEE).
  The payback period is obviously a very simple and straightforward metric, yet it does not take into account the time value for money. Therefore, it is used for simple ballpark calculations about when an organization recoups the money from an investment.
• The Accounting Rate of Return (ARR), which estimates the net profits of the investment, taking into account depreciation and the total lifetime of the maintenance project. ARR ignores the time value of money, but it takes into account the depreciation of the maintenance platforms, including relevant IT systems.
• The Net Present Value (NPV), which indicates the value of the future net cash flow of the project, taking into account the time value for money. NPV presents the total value that the investment will produce in the specified time frame. Profitable investments have a NPV that exceeds their cost.
• The Internal Rate of Return (IRR), which shows the interest rate on the capital of the investment that would yield the monetary value expected to be produced by the investment. IRR takes into account the time value of money.

Once computed, these indicators can be used to understand the value of investments, but also to compare alternative investments. For example, the NPV and the payback period provide good estimates about the monetary value of the investment in a specified time frame. At the same time, the IRR can be used to compare two or more alternative investments as a means of prioritizing them in the company’s project portfolio.

Capital Budgeting for enterprise maintenance: Costs and expected benefits

The calculation of the above-listed indicators for maintenance projects is a matter of understanding and estimating their benefits and costs. On the cost side, the following parameters should be considered:

• Software costs, including costs for software licenses and/or custom software development. In the case of predictive maintenance, multiple software packages can be relevant, including enterprise asset management systems, predictive data analytics sub-systems, data collection middleware and more.
  Given that some of these systems might be pre-existing (e.g., as part of legacy asset management), the calculation should focus on the new systems to be developed or licensed.
• Hardware infrastructure costs, including costs for data centers, computers, and networking devices (e.g., WiFi, Ethernet), as well as sensors for data collection. In most cases, hardware will be purchased up front, yet there are cases where hardware costs will be spread across more years of the period entailed in the calculation of financal indicators.
• Training and technical support costs, including costs associated with installation, training and annual support. Note that in many cases, such costs will be incurred per annum.
• Business consulting costs, including costs for establishing and justifying the predictive maintenance business cases, but also for estimating the deployment’s feasibility and profitability.
• Energy and consumables costs, including costs related to data center operation, as well as consumables, such as additional sensors and RFID tags.

Benefit estimations are much more challenging to capture. This is due to the multi-faceted benefits of predictive maintenance. There’s also the fact that complex calculations might be needed in order to translate maintenance parameters to monetary values.

In principle, benefits will stem from improvements in the following areas:

• Improved asset life and higher utilization (OEE): One of the main benefits of PdM is its ability to maximize asset lifespan and utilization. Increased utilization translates to monetary benefits, which can be significant, especially when the life and utilization of expensive assets (e.g., multi-million dollar machines) are prolonged.
  In order to associate improved OEE with a tangible dollar amount, one needs to know the increase of the asset’s utilization and the purchase price of the asset.
• Reduced downtime: Maintenance investments (such as PdM) strive to reduce asset downtime for maintenance purposes, as a result of improved planning and scheduling of maintenance processes. Reduced downtimes are linked with tangible benefits, such as less disruptions in manufacturing production, improved energy generation performance in energy plants, improved compliance to regulations (for regulated industries), higher availability of facilities and buildings, and more.
  All these benefits can then be associated with a reduction in lost revenues per annum. As a rule of thumb, one can calculate the implications of downtime on lost revenues. Accordingly, the monetary benefit from the deployment of the maintenance system can be estimated from the annual downtime reduction.
• Reduced inventory and supply chain management costs: Predictive maintenance enables better planning of available stock. In this way, it helps avoiding over-stock, ensuring improved cash flow and utilization of funds, while avoiding situations where stocked parts become obsolete.
  Likewise, it boosts avoidance of under-stock, which leads to waste of supply chain time when maintenance must take place. Overall, organizations can estimate the monetary implications of over-stock and under-stock, as well as the extent at which PdM reduces them. These reductions can be used in capital budgeting calculations.
• Reduced labor costs: Predictive maintenance systems improve labor productivity and reduce unexpected overtime. This leads to reduced labor costs, which are directly linked to the monetary benefits of PdM. In estimating labor cost reduction, one should take into account the labor hours wasted waiting for parts or rescheduling work orders when parts are not available.
• Reduced Usage of Utilities: Improvements in maintenance can also lead to better usage of utilities such as gas and electricity, as relevant units (e.g., Heat Ventilation Air Conditioning units) will have better performance and less failures. Better usage of utilities can direct lead to monetary benefits connected to reduced utility expenses paid by the organizations on an annual basis.

Mapping each of the above benefits into monetary value may involve several reasonable assumptions (e.g., regarding labor cost, purchase prices, machines utilization and more), along with complex calculations involving advanced statistics.

This is the reason why such calculations require an effective collaboration between the finance and accounting department, maintenance practitioners and engineers. Maintenance literature is full of research papers with practical formulas for quantifying the benefits of maintenance.

Limits and Challenges of using Capital Budgeting in enterprise maintenance

Calculating financial indicators such as IRR and NPV is a very useful tool for prioritizing maintenance-related investments and making educated decisions. Nevertheless, it’s very important to understand the limitations of capital budgeting methodologies:

• Uncertainty: Many calculations (especially on the benefits side) are based on educated estimates. The latter is often error-prone, which can lead to inaccurate indications. In order to alleviate risks stemming from erroneous (e.g., overly optimistic) estimations, one can consider alternative scenarios, such as a baseline estimation, as well as conservative and optimistic ones.
  Furthermore, probabilities could be assigned to each scenario, to allow for scorecard creation with alternative options that managers could use in their decision making.
• Intangible Benefits: Capital budgeting methodologies are notorious for their inability to account for intangible benefits, notably benefits that are not direct and cannot be easily quantified. Consider, for example, the improvement in employee safety during field maintenance processes.
  This is a clear benefit but cannot be quantified. It’s therefore likely that it is not captured as part of a typical ROI/IRR calculation.

Likewise, a PdM project may establish digitization infrastructures (e.g., sensor networks, automated data collection) of strategic importance, which could later enable other important projects (e.g., AR-based remote maintenance). Benefits associated with such follow-up projects are considered intangible and cannot be immediately quantified.

In the economic literature, stock options models which quantify the future right to cash out stocks are used to cater for follow up projects. Nevertheless, such models have not been widely used for industrial maintenance investments.

Despite significant efforts spent on producing ROI calculators and formulas for quantifying the benefits of digital maintenance projects, the estimation of the business value of maintenance systems (such as PdM) remains challenging.

In several cases, ROI calculation is as much art as science. No matter the methodology and tools used, it’s always good to understand how these tools work behind the scenes. We hope that we were able to shed some light on this, and is of use in your next enterprise maintenance investment-related capital budgeting assignment.