KPIs: Performance Indicators in Energy Management

The use of performance indicators to monitor processes and projects is a fundamental practice to aid decision-making, indicate opportunities for improvement and verify compliance with previously established goals.

In energy use management, it is no different. It is only possible to measure and compare the gains obtained through efficiency with the use of appropriate energy indicators.

In this context, the most common practice in the market is to use the billed energy value, focusing on the cost of the kilowatt-hour and absolute consumption figures over time (amount consumed per day, per year, per month, etc.).

In fact, this is the simplest way to control energy use. However, it is also one of the most incomplete: in this way, we lose opportunities to reduce consumption through more careful analyses.

Energy management systems, as proposed by the ISO 50,001 standard, provide for the definition of a set of energy indicators.

PDCA for Energy Management

When adopting continuous improvement through the PDCA (Plan, Do, Check, Act) cycle, the verification stage (Check) is a key activity that leads to improvement guidelines.

An analysis of inadequate indicators can lead to actions or mistakes, stagnating the process and leading to a loss of efficiency. In addition, of course, to generating tedious and unproductive work of monitoring numbers without a purpose.

This reinforces the importance of the process of choosing and customizing efficiency indicators for monitoring and improving any type of process.

There are five crucial steps to follow:

(1) Definition of the main objectives of the process or project,

(2) In-depth understanding of the activities that make up this process or project,

(3) Search for indicators already developed in the market and used in benchmarking

(4) Analysis of the data already available and which would be desired, and

(5) Creation and customization of performance indicators.

The identification of opportunities for improving energy efficiency, even empirically, goes through this process. In one of our projects, we were hired to identify these opportunities with a very clear objective: to reduce electricity costs in the client's data centers.

This presentation already brings up a reflection: how do we demonstrate the efficiency gains identified? How will the client be able to use our work in the best possible way? What decision-making process are we supporting with specialized technical support?

Case study of KPIs in data centers

Specifically regarding data centers, we know that a large part of the energy consumption comes from the cooling systems, and that the envelope of the data processing centers interferes with the thermal load inside the environment. Three performance indicators were then used.

The first allows comparison with the international benchmark (known as PUE – Power Usage Effectiveness).

The other two were customized to the client's reality: air conditioning power divided by the power of IT equipment (dimensionless), and electricity cost divided by IT power (R$/kWTI).

The indicators allowed comparison between the different buildings in which the data centers were installed, and we arrived at sometimes unexpected lessons: the operational cost of adapting conventional structures to house data centers is higher than the cost of concentrating the equipment in specialized buildings – even if this means building a new building dedicated to production processes. Key performance indicators (KPIs) end up becoming even more dependent on their main objective. After all, you can choose to measure the energy consumption of the company as a whole, of a specific production process, of a piece of equipment, of a process step, of a product, etc.

Most energy KPIs used today still have time as a reference parameter – the amount of energy consumed per day, month, or year. This approach, however, does not encompass the cause and effect relationships in energy processes. The result is difficulty in understanding and identifying savings opportunities.

Mitsidi's experience in the glass industry made it possible to define KPIs. We began the project with a detailed energy review of the glass manufacturing process. There, our assessment was prepared in 44 efficiency measures and directed as management actions to the main points of improvement.

When analyzing the available energy consumption data, we reported the need to implement a more accurate measurement system, so that customized and more precise indicators could be calculated.

This led to an extensive research study of variables for each stage of the production process. These variables began to be monitored and received indicators appropriate to their operational dynamics.

Finally, a large part of the indicators are financially standardized. According to Hamilton Ortiz, Mitsidi is working to increasingly identify how each stage of the process impacts the final energy expenditure per ton of glass produced. In this way, it will be possible to act directly on waste, increasing the competitiveness and sustainability of its products.

Conclusion

We can conclude from our experiences that choosing good metrics leads to good (and sometimes unexpected) results. Selecting a set of suitable KPIs and carefully checking the values obtained is an exciting and investigative task. It is only through this detailed analysis that we can understand and rationalize our energy use, thus achieving good Energy Management.