@PSP – Issue 2, April 2021
Performance gap analysis: inconvenient truths

 ‘Digital twin’ and  ‘BIM‘ are key words that have become almost ubiquitous in the real estate scene. In my capacity as Digital Innovation Officer, I am always keen to learn how all of these digital achievements prove their worth in practice. Could you then explain to us briefly how these are deployed for the performance gap analysis?

It is a fact that many new buildings fail to achieve targeted parameters in terms of energy consumption and comfort. There tends to be a so-called performance gap between the actual and the target characteristics. This may be due to technical reasons, or derive from simplified standardised calculation bases. Weather or climate that deviates from the simulation can also have an impact, as can occupant behaviour.

Particularly for showcase building in terms of sustainability such as the Grosspeter Tower, it is important to identify and to address this gap from an early stage. The specialist engineer developed a building model at the construction project phase. This was used to determine its future energy consumption. Together with our partners, we then created a ‘digital twin – as built’.

Initially, energy demand is simulated based on standard weather datasets. Once the building has been commissioned, however, we can use real-time weather data to compare the actual condition with. This enables us to analyse the performance gap at a high level. Thanks to the visualisation, this can be done in a highly intuitive and flexible manner.

And what did this analysis reveal?

The analysis relentlessly identifies what is not working well. This truth is not always comfortable. For example, the monitoring revealed that the required degree of thermal recovery was not being achieved by the ventilation systems. As a consequence, the effective consumption for heating and air-conditioning was about 4.5 times higher than had been calculated by the model. In the interim, the thermal recovery has been replaced.

There are also plausible deviations, however. For example, hotel water consumption was around twice as high as had been planned in the execution phase. The very high occupancy rates during the period under report broadly put this gap into context, however.

“The analysis relentlessly identifies what is not working well.”

What were the biggest challenges in this process?

There is clearly a greater engineering workload during the initial phase when it comes to setting up the data collection and trend monitoring. It is also important to draw the right conclusions from the collected data. For us as building owners, the ‘correct’ project organisation with clearly-allocated responsibilities was certainly something of a challenge. These need to be adjusted as operations progress.

In addition, the gathered meter data also needed to be verified. For example, the local installations also needed to be checked at the outset. Are we really measuring the right energy flow?

Users also have an impact on the performance of a building: how do you involve them in the process?

We need to make something clear here: The initial operational optimisation, performed immediately after the building has been handed over, focuses on making sure the occupants feel comfortable. The actual energy optimisation is performed at a later stage. I can’t deny that there are certain potential conflicts of interest here which can be challenging.

Are you talking about feeling comfortable in the winter months with a set room temperature of 21°C? I read somewhere that these guidelines are based on a male standard. No wonder the women feel cold…

I think it’s described as a 70 kg ‘standard person’. But indeed: only a small proportion of the occupants effectively correspond to the standard person. The personal sensation of comfort is a correspondingly individual parameter. Room air temperature, surface temperature and the characteristics of the surrounding surfaces or desk lighting – all these factors have an impact. Such conflicts of interest can be defused to a certain extent by recommending the ideal furnishings and raising awareness of the relevant issues from an early stage. However, we cannot dictate to people how they should feel comfortable. We want to make sure occupants feel at ease in our buildings and are able to work productively – that must be our goal.

“One cannot place the blame for the performance gap with the occupants.”

But I would like to stress one thing here: One cannot place the blame for the performance gap with the occupants. The principal cause of the performance gap – at least in a building such as Grosspeter – is the complexity. The technology deployed is not intrinsically ‘rocket science’. The challenge lies in the way that systems are linked together and are dependent upon each other.

Is a digital model an absolute precondition for the performance gap analysis?

In a new building, I would say a sound energetic operational optimisation and performance gap analysis are not possible anymore without a corresponding building model. This is the only way to simulate the effective energy demand over the entire annual cycle in a realistic manner, and to stipulate this as early as the project phase. All systems are configured and dimensioned on the basis of this.

In the case of this project we deliberately chose to maintain the model with the corresponding evaluations. This means the simulation model can continue to be used in various ways during the operating phase. If material changes of use occur, then the updated model can quickly assess whether the changed energy demand is due to an inefficiency in the system or to the change of use.

However, it is not absolutely necessary to keep updating a model at this level on an ongoing basis. A good building management system with measurement topology that focuses on energy flows together with corresponding trend logging supplies the most important tool for operational optimisation.

So you don’t see the digital twin as an integral part of the operational phase yet?

For us, the pure digital twin as a model is currently not part of the operational phase. In our properties, ‘standard operation’ seldom persists over an extended period of time. In larger buildings, in particular, there are always changes happening somewhere, or tenant fit-outs are being undertaken. Operations therefore have to be constantly adjusted and optimised. As a consequence, these adjustments would have to be reflected in the digital twin model on an ongoing basis. In practice at the moment, with project teams operating in different ways, this cannot yet be implemented in a beneficial manner. Clearly, when it comes to meeting the comfort requirements of our tenants, while at the same time achieving our sustainability goals, we are dependent upon data and suitable tools – but we don’t need a fancy model in every case. What is more important, it seems to me, is to consciously collect and request data and information, and then put this data to use. We have been doing this actively and in an increasingly targeted way for some time now.