Gap-Filling Methodology

Throughout the Real Estate sector, consumption data may not be available for all assets or even within assets. Where there are gaps in consumption data for a standing investment asset, SIERA+ aims to provide a reasonable estimate.

Gap-filling missing data is less accurate than using complete data but allows users to see an indicative, realistic picture of a fund or asset's performance. Whilst the approach aligns with the best practice where applicable, there may be legitimate deviations from other reporting outputs depending on the application and purpose of the deliverable.

For example, GRESB limits how much data can be estimated by 20% of the total consumption value or three months within a single 12-month reporting year. Similarly, specific carbon footprinting and reporting exercises might yield different results due to emissions factors and specific benchmarks used.

EVORA has developed a multi-stage approach in SIERA+ to estimate missing data. This approach considers the length of gaps and other available data. The approach fills gaps for time-based completeness (days missing) and area-based coverage (where asset floor area is not covered/reported). The method is applied at the month/meter level to ensure the most accurate estimates when aggregated up to the asset and fund level.

Estimates generated by SIERA+ are limited to the analysis presented in the reporting capabilities in the user interface and do not substitute actual data reported. For the above reasons, they are not used automatically in reporting submissions such as GRESB.

Two main types of gap-filling are used in SIERA+: area and time.  As a result, two approaches are required to indicate an asset and fund's performance fully.

Area Based Gaps

Area gap-filling is used when consumption is not reported for the whole Asset. 

For example, when data is available for common areas but needs to be added for specific tenant areas, SIERA+ follows the thought process outlined below to address this issue and ensure comprehensive data analysis.

  There are a couple of additional notes to beware of with area-based gap-filling:

• Where GIA is not reported for an asset, GIA is estimated using asset class-specific benchmarks based on a ratio of the net lettable area (NLA) to GIA, aligned to those used by GRESB.
• If an asset is known not to have a specific utility type used on site ( e.g. gas), then no estimates will be applied for that utility.
• Whilst gap-filling supports the reporting of full building area coverage, the threshold for assuming 100% area coverage for gas consumption is total area coverage >= total net lettable area. This is due to the everyday challenges in unmetered space heating where assets report gas as shared services but represent whole building coverage. 
• Outdoor/exterior parking meters are excluded from area-based coverage gap-filling calculations.

Time-Based Gaps

Time gap-filling is used when a meter may be set up on SIERA+. However, it may only report consumption for a few months.

Incomplete data shouldn't stand in the way of getting what needs to be done, so SIERA+ employs the process depicted in the following diagram to provide users with accurate, comprehensive information.

Examples:

Method 1

A meter reports 1000 kWh between 1st May and 16th May. This is equivalent to 62.5 kWh/day.  Within this month, there are 15 days of missing data, 15*62.5 =937.5 kWh of gap-filled data.

Method 2

A meter reports continuous consumption data from January 2021 – September 2022, with no consumption reported in October – December 2022.

January – September 2021 reported 500,000 kWh, whereas January – 2022 reported 510,000 kWh. This is an increase of 2%.  Consumption reported for October – December 2021 was 150,000 kWh.  One hundred fifty thousand increased by 2% would lead to 153,000 kWh of gap-filled data.

Method 3

A meter only has data from 1st January 2022 to 30th September 2022 and records 100,000 kWh. This is an average of 366 kWh/day.  To get an indicative value for all of 2022, there are 92 days with no consumption reported from October to December.

92*366 = 33,700 kWh gap-filled data.

Using benchmarks for time-based gaps.

A meter serves 100m² but only has consumption data for January 2022. Therefore, within all of 2022, this meter is missing 334 days.

An industry standard benchmark intensity rate for the relevant sector and utility type is 150 kWh/ m2/ year. This is equivalent to 0.41 kWh/m²/day.

= (Relevant daily benchmark figure * Area assigned to the asset * Number of days missing)

= (0.41 * 100 * 334) = 13,726 kWh of gap-filled data.

Time-based estimates will only be calculated within the ownership dates of the asset or if the meter is only known to be operating for a specific time frame.

Additionally, if a consumption value of "0" is entered, it will not be considered a gap. SIERA+ will only fill in the missing data where there are null values.   

Universal Similarities

As stated above, both methods are two sides of the same coin, so naturally, some parts of the methodology are unanimous. They are as follows:

• Default location-based carbon conversion factors per utility/country are used if the gap-filled data relates to an area-based gap.
• Benchmarks are proprietary for consumption (energy and water) gap-filling and are based on a whole building model by Asset class and country where available.
• If there is no industry-standard benchmark for that country and property sub-type, then we will reduce the granularity of the benchmark to the next available alternative.
• Gap-filling methods 1 and 2 indirectly consider seasonal variation as they are extrapolating based on historical data from relevant periods.