Larsson (1993, 1995) via Papamichael (2000) says that "the design of buildings requires association of all building disciplines and involved parties to formulate the initial phases of building design". It is essential that this alliance remains apparent throughout the entire life-cycles of the building from conceptualisation to demolition, in order to produce a holistic design that is easy on the environment and can be labelled as a true 'green' building. Resolutions taken at the design stages reflect in the final production and performance of any building and that in turn affects its rating (Papamichael, 2000).
Credibility gaps have become part of the functioning of a building post-construction. It refers to the gap between the design intent and the actual performance of the building. The gap arises
"not so much because predictive techniques 'wrong', but because the assumptions often used are not well enough informed by what really happens in practice, because few people who design buildings go on to monitor their performance." (Bordass et al, 2004)
Most common of the reasons for this gap are poor building commissioning, poor briefing, unrealistic design assumptions, inefficient management and unaccountable user behaviour (Bordass et al, 2004).
Currently, standards and tools that are used to evaluate one part of the building life-cycle is not applicable to some other part. This makes it immensely difficult to compare targets that are set at the inception, acquisition or the design stage of the project with the actual operation and performance of building once it has been constructed. Hence, there is no proper relevance or harmony between each aspect of a building which automatically gives birth to a credibility gap and in turn the building does not perform as a true 'green' building in reality (UKGBC Code for Sustainable Buildings Task Group).
Users and how they use the building have a huge impact on the building's CO2 emissions, waste production and its overall performance as a 'green' building. This almost always leads to different performance than the actual predictions by the rating tool. Accurate building energy simulations, calculations and life-cycle analyses procedures may predict a certain high rating and yet the building may not perform accordingly in reality. This leads to an issue of whether we are actually making 'green' buildings which battle climate change. Carrying out assessment post-occupation assures that the rating awarded is a true picture of how the building actually performs in reality (Bordass & Leaman, 2004).
It is interesting to note, from , that BREEAM assesses a building in its design and post-construction stages and the assessment method includes a mixture of performance-based and feature-specific criteria. There is no post-occupancy assessment as such, unless the building already exists and is being assessed under BREEAM In-use (BRE, 2010). Hence, any design decisions will reflect on the achieved rating which can make or break the building's business case and popularity as a 'green' building. To assess a building, BREEAM employs energy simulation software and actual consumption figures from utility bills if available. However, even with a high amount of accuracy and intricate computational details, there is always a gap between the predicted rating and actual building performance (Papamichael, 2000).
A good example of post-occupancy evaluation highlighting this gap is a study conducted by Curwell et al in 1999, included a post-occupancy evaluation of an award-winning office building two years after its completion.
The graph () compares all the gas and electricity predictions made by BREEAM, the design team and ECON 19. The numbers show that the building produced annual CO2 emissions just below 40kg/m2 whereas the estimated BREEAM emissions were just over 40kg/m2. The actual amount exceeds the predicted amount three times over (Bordass, 1999 via Hes, 2007). What is even more disturbing is that the BREEAM prediction is even lesser than what the design team predicted; there is a difference of slightly over 10kg/m2.
Annual CO2 emissions (kg/m2 treated floor area)
Figure : Annual CO2 emissions from operational energy use in an environmental award-winning head office building complex in UK (Source: Curwell et al, 1999, modified by Author)
There is a lot of research related to identifying this gap. Much of it seems to relate to user interactions and control of the building environment, and that these are not identified or modelled in the predictions made during the design stages. Discrepancies of up to a fact of 2 can occur. These discrepancies can be reduced (and are being addressed) by modelling behaviour in our building simulation tools. The other key area that causes the gap between prediction and performance is that the design assumptions are optimistic (e.g. related to air tightness) and in reality the performance assumptions are either too difficult to deliver on site or changes are made in the design on site resulting in a different performance (Steemers, 2010).
This is supported by the case above as pointed out by Bordass et al (1999). Inaccuracies in assumptions during initial estimations as well as inaccuracies during design, construction and commissioning stages may have led to such inconsistencies in actual building performance. The major reason the design and BREEAM predictions were highly inaccurate was that it did not include electricity consumption by computer/communications equipment, HVAC systems and underestimated calculations of energy consumption by equipments when not in use. Lack of optimum performance of the HVAC system was also a contributing factor. The ECON 19 prediction however, did consider these issues (Bordass et al, 1999).
A second study conducted more recently by Sawyer et al (2008), included observations on two similar buildings in the UK, one of which is a BREEAM assessed building with 'Excellent' rating (Building A). The table below compares the performances of the two buildings (), and then measures them against ECON 19 benchmarks ().
Table : Metered energy use at buildings A and B (Source: Sawyer, 2008)
Table : Energy consumption for the buildings compared with ECON19 benchmarks (Source: Sawyer, 2008 modified by Author)
It is evident from the tables, that compared to the benchmarks, building A performs just as a "Typical" structure as building B does, which shouldn't be case considering that building A is rated 'Excellent'. In fact, from it seems that building B's total energy consumption per year is closer as a value to ECON19's "Good Practice" benchmark. The study concludes that "achieving a BREEAM Excellent rating does not appear to guarantee a good post-occupancy energy performance". (Sawyer et al, 2008).
Assessing at only design and post-completion could possibly be a problem that BREEAM faces. However, Steemers (2010) argues that this may not necessarily be a shortcoming, because a lot depends on the accuracy of the building model. "A greater rigour with respect to the variables and assumptions, and an assessment of the 'robustness' of a design would be useful. Nonetheless, this is never carried out, leaving the inputs to the assessments open to misuse or too optimistic." However, he also adds that over time rating systems are adapting post-occupancy assessment to increase the viability of the system, and it would be beneficial to have it this assessment as "an on-going one rather than a one-off".
To assess whether a green building is really 'green', with minimum carbon footprint and to know its actual impact on the environment, it is necessary to carry out assessment into post-occupancy.
An identical problem can be spotted in the implementation of USGBC's LEED.
The LEED system has changed the market for environmentally friendly buildings in the US. However, the best data available shows that on average, they use more energy than comparable buildings. What has essentially been created is the image of energy efficient buildings, but not actual energy efficiency (Gifford, 2010).
As observed in case of BREEAM, poor building modelling, absence of post-occupancy assessment and user behaviour can lead to widening the gap between performance prediction and actual building performance (Bordass et al, 1999). LEED uses either Prescriptive Compliance Path (allowing the project to achieve points based on prescriptive measures laid down by ASHRAE standards) or the Whole Building Energy Simulation option (where modelling is used to predict consumption/emission) to assess building energy performance (Roderick, McEwan, Wheatley, & Alonso, 2009). For LEED, while modelling a building, energy simulations of both base building and design building are performed, from which the annual energy costs of the building is determined along with percentage savings on heating, cooling, lighting and power loads (LEED-CS, 2009). However, perhaps detailed and accurate energy modelling is needed to make sure more buildings perform according to the rating awarded to it.
Experts in the field disagree to a certain extent. Schendler (2010) argues that, modelling is most valuable as a design tool, and intelligent modelling early in the design process is more beneficial and efficient than a complicated model. Gifford (2010) supports this by saying that more modelling is not the answer. According to him, buildings are simply too complicated to be modelled accurately, even if the model perfectly reflects the building. Building designs don't reflect too much heat or cooling delivered to different rooms, gaps in construction, or any other defects. Building designs are usually very incomplete, and the model can't be more complete than the design. Modelling can never work. Better predictions of energy use can be made by comparing energy input to heating, ventilation, and cooling systems at peak to building size.
However, on the question of whether a rating assessment should carry on to post-occupation, Gifford has strong views on the benefits of post-occupancy assessment. He emphasizes that public scrutiny of utility bills for all rated buildings is important. Rating buildings according to actual energy consumption can a rating system reward success, and encourage energy saving in not only the design phase, but also during construction, as well as after the building is occupied (Gifford, 2008).
Figure : Average Scores: Green Vs. Rest of database (Source: Huizenga et al, 2005)
A survey was conducted by Huizenga et al in 2005 on occupant satisfaction both in LEED rated and non-LEED buildings (from a set database of buildings) (refer to ). The average scores above show that most LEED and self-proclaimed green buildings have all received 'Very Satisfied' votes. However, so have 'non-green' buildings. Also, in , it can be seen that though most 'green' buildings lie beyond the green median, there is a certain percentage of 'green' buildings that lie within the non-green median. From Figure 20 it is concluded that 10% of non-LEED meet the criteria of 80% or more occupants satisfied with thermal comfort, and about 20% of LEED rated buildings satisfy the same criteria. No doubt, occupants in more and more LEED rated buildings are happy with the building design and IEQ (Huizenga et al, 2005).
Figure : Occupant Satisfaction of Building Overall (Source: Huizenga et al, 2008)
Figure : Thermal Comfort: Occupant Satisfaction (Source: Huizenga et al, 2008 modified by Author)
However, no clear relationship was found between LEED credit points and occupant satisfaction with IEQ (Huizenga, 2008) and the question remains: out of the numerous LEED buildings studied, why is it that only 20% of the criterion is achieved? They are only slightly better than non-green buildings and more importantly, is that 20% enough to reduce CO2 emissions and carbon footprints of buildings by a considerable amount?
LEED looks at overall energy reduction as an indicator and award points for it (Ng, 2009). With the recent exponential growth in the number of LEED certified buildings, the number of occupied LEED buildings can be studied to find out how measured performance meets energy efficiency objectives. Turner and Frankel's study (2008) of energy performance of two LEED certified buildings (New Construction) study comes up with interesting findings. It is generally assumed that the performance baseline LEED adapts to make buildings perform significantly higher than the baseline defined by CBECS. However, the buildings chosen for the study produced average performance close to what the national building stock produces (Turner & Frankel, 2008).
Buildings generating excess energy than what was predicted
Figure : Actual/Predicted Ratios Relative for Design EUI (Source: Turner & Frankel, 2008 modified by Author)
pits Ratio of Measured/Design EUI against Design EUI, where the value of '1.0' on the x-axis denotes that a building's actual performance equals predicted performance. The scatter on the graph includes projects that go almost up to '2.5', denoting that the projects use almost 2.5 times the amount of energy predicted by the rating. The vice-versa is also true, where projects go down till '0.5', denoting that their energy use is half of that predicted. The actual energy performance of more than half the buildings in the study, digress from their predicted energy performance by more than 25%. Only 30% performs notably better and 25% performs notably worse (Turner & Frankel, 2008). As a result, contrary to practices promoted by BREEAM, the report suggests how, although energy modelling is a good gauge of performance outcome, its results could be highly inaccurate and diverge extensively from actual building performance. The study goes on to suggest the usage of life-cycle cost estimation for buildings for more accurate predictions and that more response and follow-up is needed from actual building performance results to improve future building modelling and benchmarking (Turner & Frankel, 2008).
However, the newest version of LEED (LEED 2009) includes Minimum Program Requirements, which are in addition to the pre-requisites that LEED already prescribes. This is a step forward to monitor this gap in performance. The MPR (USGBC, 2009) guide specifies,
"Sharing this data includes supplying information on a regular basis in a free, accessible, and secure online tool or, if necessary, taking any action to authorize the collection of information directly from service or utility providers. This commitment must carry forward if the building or space changes ownership or lessee."
However, LEED is now demanding post-occupancy performance data from certified buildings and therein lies another significant difference between LEED and BREEAM. Beginning April 2009, all new projects registered under LEED 2009 are required to report monthly energy and water data for a period of 5 years, every year (Ng, 2009). There has also been an effort to demand the same from conventional uncertified buildings with the cities of Washington, DC; Austin, TX; New York, NY; Seattle, WA; and Portland, OR; and the State of California independently implementing legislation for disclosure of building energy efficiency (mostly for commercial buildings). Additionally, USGBC's recently introduced Building Performance Initiative which will collect and analyse post-occupancy data from existing LEED certified buildings is another significant step towards closing the 'prediction vs. performance' gap (LEED Chicago Chapter, 2009).
3.4 Green Star
A relatively new rating tool, Green Star, like BREEAM and LEED is also has no post-occupancy assessment. However, independent studies conducted on Green Star rated buildings have had positive conclusions. A major challenge for Australian emission reductions is its existing buildings. According to McKinsey (look up), the building sector offers 60 million tonnes of carbon reduction abatement opportunities by 2030, the majority of which are able to be implemented today. Currently, the GBCA is working with the Government and industry to address this problem. Tenant education is another huge area that needs more resources and time and the gains of a 6 Green Star rated building can be undone by tenant operations (Pemberton, 2009).
The first building in Australia to receive a 6 Star rating ('World Leadership') was the Szencorp Building in South Melbourne. Thomas and Vandenberg's study (2007) of its building performance and post-occupancy evaluations revealed interesting results. The table below is the collection of data that compares what the initial design aimed at to what the actual results were at the time of study (Thomas & Vandenberg, 2007).
Figure : Actual building performance against design intent: Szencorp (Source: Thomas & Vandenberg, 2007)
The office was refurbished and re-opened in the year 2006; and it won a 6-Star rating under Green Star Office Design and a post-occupancy rating of 5 stars under NABERS Office Water. In the study the following observations were made (refer to ). It can be observed that Gas consumption of the building exceeded the design intent by almost 200GJ, the chunk of it taken up by combined heating and cooling, during the time period from December 2005 to November 2006. On the other hand, use of heat pumps and on-site generation facilities, electricity consumption was lesser than predicted - not significantly, but by about 10MW-hrs. This is reflected in above.
Figure : Gas and Electricity consumption of Sczencorp, from Dec '05 to Nov '06 (Source: Thomas & Vandenberg, 2007)
The Szencorp Building Performance Report of 2009, however, shows drastic inconsistencies in gas and electricity consumption of the building (refer to ). The Gas Consumption Predictions that were made hasn't proved to be accurate and the fluctuations in the amounts during the following years, 2006-2009, have been broad and conflicting by 30% each time. The same has been observed in case of electricity consumption, although the consumption amount is still clearly within predicted limits (Szencorp BPR, 2009).
Figure : Gas and Electricity consumption from 2006 to 2009: Szencorp (Source: Szencorp BPR, 2009)
Conversely, post-occupancy evaluation on the building revealed that the energy performance of the building had nothing to do with indoor environmental quality. In the IEQ area, the study confirms that the building functions really well as all users are highly satisfied with indoor features. It is likely that energy consumption only affects the cost cover of operating and maintaining the building (Vandenberg, 2009).
The review highlights how a building is still considered to be a 6-star Green Star rated building, even when it does not perform as projected in terms of one of the most important areas of performance: energy. A well thought-out briefing process, constant post-occupancy assessment and benchmarking, and continuous receptive building management during both commissioning and operation more than just a trial period of 12 months, is necessary to monitor successful building performance and help increase occupant productivity (Thomas & Vandenberg, 2007).
"A total of $6.12 million was spent on design and building development for 40 Albert Road (Szencorp Building). The client believes the money and effort expended would be difficult to justify for a single building. However, they also believe the investment has more than paid for itself in terms of the learnings, the profile the project has received, and the ability of the owner to develop a new level of business services in the rapidly growing market of leading-edge green buildings." - Thomas & Vandenberg, 2007.
On questioning whether there should be more stress on building modelling, Chapa (2010) said,
"The performance of a building model can, and should, only be used to compare the potential performance of that building against another building's model with the same set of assumptions (or as close as possible). Modelling should be used as a shorthand of the building's predicted actual performance, not as a measure of the building's actual energy use."
The Ministry of Environment, New Zealand suggests the use of Integrated -Whole Building Design Process. Adopting the process from the start of the project through to the commissioning, operations and on to post-occupancy stages, would be beneficial, since the process considers the building as a whole and requires the involvement of all stakeholders, design team members and future users or tenants. The rating tool can be used at the end of every stage to monitor and evaluate the success. However, it is necessary to avoid using the tools during the design process since this might lead to point-mongering.
Australia, however, uses NABERS to evaluate post-occupancy performance of buildings. There are discussions going on to link NABERS with Green Star for a total assessment system, which would definitely help energy savings (Chapa, 2010)
3.5 Green Mark
One of Green Mark's primary assessment necessities are its post-occupancy conditions that apply to all building type; most important being that the rating for any Green mark building is valid for only for a period of three years. The project has to be re-assessed every three years after its completion in order to maintain its original rating. This literally forces the building to keep up its performance according to its certification standards (the minimum being 50 points; Green Mark certified). Hence, all Green Mark buildings by default save 10-15% of energy consumption minimum. Constant monitoring and commissioning by, developers and owners hence becomes a necessity; the need to update the building taking care of filling gaps caused by user behaviour, people-hours over-occupancy, equipments failure and so on, become important and a must to regain the same rating three years later (Foo, 2005). The Green Mark terms and Conditions Guide states this clause as:
"BCA's Assessment of the Project, and the information and opinion contained in the Certificate or Report shall be valid for a period of 3 years. Nothing in the Certificate or Report shall be taken as warranting or guaranteeing that the environmental performance of the Project will remain in the condition as stated in this Certificate or Report as design changes, building additions and alterations, misuse and accident may occur after the Assessment. All implied terms and warranties are expressly excluded to the maximum extent permitted by law." (BCA, 2008).
Considering the steep target set, this clause was put down to ensure that buildings continue to be well maintained and managed post-construction and post-occupancy (Foo, 2005). In case of existing buildings, this clause is optional if either
The building demonstrates performance of having achieved 10% energy savings over the period of the last three years, or if
The EEI of the building lies within 215kWh/m2/year for offices, 420kWh/m2/year for hotels or 479kWh/m2/year for retail developments.
Moreover, BCA has tightened the noose on existing buildings by putting in place the condition that requires the energy savings of the building to be 15% after its certification (BCA, 2008).
This is all in addition to exclusive credits awarded by BCA just to conduct post-occupancy evaluation taking into consideration technical, energy and environmental performance, financial effectiveness and occupant health and satisfaction and continuing on to corrective actions (BCA, 2008).
BCA authorities confirm that till date there has been no cases of 'under-performance'. Considering that Green Mark was introduced in 2005; buildings that have been certified by Green Mark, the numbers are still in the hundreds. Moreover, only a certain percentage has undergone re-assessment to confirm their original certification. In a span of five years, there hasn't been any case where a building failed to perform according to its design intent. Case studies on such buildings are non-existent (Keng, 2010). However, it should be kept in mind that further re-assessment of building down a few more years may reveal credibility gaps and similar problems faced by the other rating systems, considering that buildings will always be deteriorating to small extents over time.
From a review of the Green Mark Assessment Guides (v1.0 and v3.0), it is clear that a high premium is put on building modelling, the ETTV & RTTV methods to building performance prediction and heavy stress on post-occupancy evaluations and rigorous commissioning.
It is interesting to note how the different rating tools have their own preferences on evaluation. While BREEAM and Green Mark stress on building modelling, LEED and Green Star prefer incorporating proven prescribed methods by building standards.
3.6 Life-Cycle Assessment
"Life-cycle assessment is a "cradle-to-grave" approach for assessing industrial systems." (SAIC, 2006)
LCA considers the whole life-cycle of a product, in this case a building project, starting from its design, material procurement, construction, handover, operations, maintenance and demolition. The impact of all these factors is put together to evaluate the impact the building will have on the environment (Curran, 1996, SAIC, 2006). A schematic diagram () shows the skeletal process of LCA; how raw materials and energy go into manufacturing and result in various outputs.
Green building rating tools, especially the ones chosen for this study, deal with similar assessment criteria (site selection, energy, water efficiency, indoor air quality, waste management, etc), and hence the demand for sustainably procured materials and sustainable transportation services be a part of the integrated whole building design. Integrating LCA tools into this would mean a better understanding of the building's impact on the environment, economy and occupant productivity; and a much more accurate building performance prediction (Trusty & Horst, 2006).
The two major steps in a life-cycle assessment are:
The inventory stage where the whole list of emissions that will occur and all the raw materials that will be used during the life-cycle of the building, is drawn up,
Evaluation of the impact of these emissions and raw materials are calculated (Cullen, 1999).
LCA needs detailed analysis of the full inventory and careful calculations in an organised manner. In addition, the entire process is time-consuming and open to high risks of inaccuracy (Cullen, 1999).
Figure : Life Cycle Stages (Source: EPA via SAIC, 2006)
LEED v3.0 incorporates LCA into its assessment of buildings to facilitate integrated design and ensure efficient environmental performance considering the entire life-cycle of the building. It covers a number of 'Materials' credits. However, while LCA is helping individual credits to an extent to produce specific solution, the overall results of building performance are very inconsistent and inaccurate. Refinement in this case is essential (Scheuer & Koeleian, 2002). LCA is also used to support BREEAM and the CSH in their certification process. This was enforced with the release of the Green Guide to Specifications by the BRE in June 2008. The specifications encompass BRE Global Environmental Profiles Scheme for Type III EPDs for construction products (BRE, 2010). The GBCA also recognizes LCA as an extremely helpful aid. However, it hasn't been incorporated into the system, the primary reason being that LCA tools currently in practice are not adapted to 'Australian-specific' Life-cycle Inventory. Nonetheless, GBCA is thoroughly exploring the prospects of incorporating LCA tools in future versions of Green Star (GBCA, 2010). Green Mark does encourage implementation of LCA tools and uses the criteria of 'Other Green Features' to award the use of these tools, however, it is not a necessary evaluation step to be taken considering that information about LCA is still insufficient and in development (Keng, 2010).
Table 9 gives a summary of observations from the chapter.
GBCA Green Star
BCA Green Mark
(from case studies)
Energy consumption exceeds by a factor of
0.9 to 2.5
Energy consumption exceeds by a factor of
0.25 to 2.5
Energy consumption exceeds by a factor of
0.2 to 0.6
No cases exist where actual performance has deviated from predicted performance
No Assessment or Evaluation
Owners have to produce energy performance data annually for 5 years after completion
Building Tuning is required for a period of 1 year
No Assessment or Evaluation
Public sector buildings need to display EEBC
Mandatory re-Assessment required every 3 years to maintain rating
Preferred Design Process
Passive Design Methods
Stress on building modelling
Stress on prescriptive outlines by building standards
Integrated Whole-Building Design
ETTV and RTTV used for energy calculation.
Building Modelling is a minimum requirement for Gold Plus and Platinum rating
Use of Life-Cycle Assessment Tools
Uses LCA to a certain extent
LCA has been included in the updated LEED 2009 (v3.0)
Not required to use LCA yet
'Innovation' credits are awarded if LCA is used
Not required to use of LCA yet
'Innovation' credits and 'Other Green features' credits are awarded if LCA is used
Table : Summary of Energy Performance