ESSAY 1
Building performance evaluation
Richard Partington, Studio Partington
At every stage of building design we have to navigate a multitude of technical standards and regulations, from the mundane rules that fix the width of a stair to the elusive performance indicators that seek to safeguard ‘design quality’. But, as Bill Gething explains in his essay, despite this barrage of performance standards, the difference between what we intend, through design, and what we achieve, in reality, is widening – most obviously in energy use, but also in comfort and productivity.
Although evidence of a performance gap has existed for decades, it is only recently that industry has widely recognised that what matters is the building’s actual operation in use. A new stage in the restructured RIBA Plan of Work (Stage 07) anticipates an architect’s involvement post-completion. Though loosely defined it recognises that information gathering, evaluation and review after handover will help to get buildings working better. However, it is not just the immediate period of ‘bedding in’ – when the systems can be monitored, adjusted or fine-tuned – that should be of concern. If we really want to understand how buildings are working and improve our future designs, we need to commit to a sustained period of observation and some form of post-occupancy evaluation (POE).
This essay introduces the common tools for structured information gathering (from the indicative to the more diagnostic), and suggests some of the benefits for architects and their clients.
Types of evaluation
The techniques for POE are well established. Building on the pioneering work of the Building Performance Research Unit at the University of Strathclyde in the 1960s and 1970s, the PROBE Studies1 of the 1990s combined an assessment of technical and energy performance, an air-pressure test on the building fabric, and a user survey based on a concise and targeted questionnaire.
Similar tools had also been applied in the United States in the study of university dormitory buildings, housing and Federal buildings. Wolfgang Preiser, who documented the evolution of POE and developed process models for conducting evaluations, describes POE as ‘a process of systematic performance evaluation of buildings after they have been built and occupied for some time’.2
A POE can serve a range of purposes from simply observing a building in use ‘as it is’ to testing against specific performance criteria set in the briefing process, but it always has a component of user feedback. POE tends to be undertaken in the early years and ideally provides useful information for the client, design and building team. The term building performance evaluation (BPE) usually refers to the whole field of building study including purely academic research. More recently work by the Usable Buildings Trust has translated these techniques into more client-oriented tools, helping to diagnose and solve problems and get buildings working efficiently for their users, balancing ‘hard’ technical measurement (temperature, air quality, carbon emissions) with ‘softer’ issues (productivity, ease of use, comfort) identified by questionnaire, interview and observation.
Soft Landings
POEs show that new buildings seldom work at their best without some running in, like the ‘sea trials’ of an ocean vessel. However, as Bill Bordass observes, most designers are only required to direct their efforts to design and construction up to handover: ‘their job description being to procure the facility, not to breathe life into it’.3
Soft Landings4 was therefore developed to help improve the situation, using a framework that can be applied to any procurement process, to give more attention to in-use outcomes. It allows considerable flexibility, so that teams can make it their own. It starts at inception and briefing, helping to change the culture of a project. It continues with review and reality checking through design and construction, before entering the critical stages in the weeks or months immediately before and after handover and finishing with extended aftercare and POE. Experience, particularly though the User Group convened by BSRIA, shows that the most important ingredient is leadership, helping team members to collaborate and understand each other.
Most buildings do undergo some testing: an airtightness test and possibly some acoustic testing, and commissioning of heating and ventilation equipment. But these tests are usually done once, to demonstrate regulatory compliance, rather than seasonally to prove effective operation throughout the year. Rigorous commissioning of equipment and controls is essential, but is no guarantee that a building will be comfortable to occupy, practical to operate and maintain, or perform as intended. An essential component of a POE, therefore, is the user’s assessment, and this implies a judgement of successes and failures.
Recent studies
BPE has continued since the PROBE studies, spurred by a greater adoption of the Soft Landings methodology. In addition, work on domestic housing in the UK and internationally began to reveal a performance gap between design intentions and energy in use. The government-sponsored BPE studies, funded by Innovate UK covered buildings of all types and combined the three common POE activities: a review of the design intention, measurement and data collection, and a survey of user experiences. For the user survey the BUS (Building Use Studies)5 methodology was adopted, and for reporting energy use CIBSE’s TM 22 tool6 was required for all non-domestic projects, with the aim being to make ready comparisons across each building type. Innovate UK also required data from the projects to be uploaded to CarbonBuzz7 and that the studies provided feedback to the occupier of the building. The Innovate UK projects gathered lots of measured information over a long period, often frequently. One of the aims was to build a better picture of how new buildings were performing for benchmarking, but it is not necessary to collect large amounts of metered data to conduct a useful evaluation. The outcomes of this study are discussed in Essay 2.
Levels of evaluation
It is possible to undertake POE for individual projects without getting research funding, but one needs to establish the scope and purpose of an evaluation by setting reasonable expectations and working to a manageable programme, not collecting data for the sake of it. The resources for POE must be considered early on and built in to the relevant work and tender stages to ensure that everyone involved is committed. Preiser emphasised the planning and briefing as well as the actual evaluation, and the RIBA has sensibly coupled Stages 0, 1 and 7 in its guide to the Plan of Work.8 He describes three distinct levels of evaluation: indicative, investigative and diagnostic.
Indicative studies based on walk-rounds and user feedback give an indication of the buildings’ strengths and weaknesses and may trigger improvements in themselves, especially where they reveal difficulties with the use of controls and management systems. Indicative studies may also shape a second phase of investigation. Learning that a building is too hot in summer might prompt investigative study of internal and external temperature measurements, CO2 levels, equipment gains and occupancy. Generally these will be objective measurements to obtain a better understanding of causes and effects and to help solve problems. In theory some or all of this information will be available from the building management system or the sub-meters, if installed and commissioned correctly.
Further study might be at a component level (heat pump or fan efficiencies) or in the detail of the controls software and settings, leading to improvements informed by the previous investigations rather than by hopeful experimentation or ‘fiddling’ with the controls. Diagnostic POE aims to thoroughly correlate environmental measurement with subjective user response to establish what can be done to improve a building and provide learning for the future. In Retrofit for Purpose9 Roderic Bunn gives a full description of BPE processes and illustrates how evaluation activities can be integrated within the Plan of Work. The main principles apply to both new build and retrofit, and general advice is to take on a manageable amount without being too ambitious.
- Adopt handover procedures that recognise the time the building will take to ‘bed in’.
- Start POE with a structured walk-round with an experienced evaluator, the contractor and the building users.
- Undertake a user survey and an energy audit.
- Use measurement and analysis in a graduated way (observe, investigate, interrogate) based on the performance measures that are important to the user.
- Use Soft Landings to change cultures and aspirations, not as a bolted-on box-ticking exercise.
The benefits of evaluation
In addition to being used to optimise performance and support building users, evaluation can also be used to set design targets within projects and to learn and provide feedback to practices across projects. Many practices in this book are using evaluation as a tool to deliver better quality to clients through these processes.
For example, BPE has played a substantial role in setting the environmental targets and design standards for the Joseph Rowntree Housing Trust (JRHT), which has been working on plans for a new community at Derwenthorpe on the edge of York for over two decades. Derwenthorpe aims to be sustainable, equitable and affordable: a template for future developments. Evaluation of the performance gap and the wider sustainability objectives has been built in to a feedback process as the project has developed, through a series of prototypes and in the delivery of the main project (see Case Study 9).
The idea of the prototypes was to develop workable construction details and heating and ventilation systems using affordable technology, so that all the future homes could adopt these on the large-scale build (550 homes in total).
The learning gleaned from the prototypes was transferred via JRHT’s team to the development partner, who adopted the construction details developed for the prototypes. A robust feedback and learning process was embedded in the early building phases, and the results of testing set a new benchmark for volume builders. All of the Phase 1 homes achieved an airtightness of around 2m3/(h.m2) with very little variation across the phase – an indication of the consistently high build quality. Testing, measurement and inspection also played a significant part in the delivery of the first phase.
Not every project sets out with ambitions to influence industry, and not every client would be willing, as JRHT has, to share its findings, including the shortcomings of earlier pilot projects. However, the work sponsored by JRHT has been highly influential, initiating research into the performance gap and helping develop techniques for measuring housing performance. For larger institutional clients who are building over a long period, benchmarking and performance testing are recognised tools for improving the quality of their estate over time – resulting in long-term savings on energy, and more productive and happier occupants.
Evaluation can also provide feedback across different projects. The Woodland Trust project (Case Study 1) builds on work from an earlier project for the National Trust, at its central office, Heelis in Swindon. The Wilkinson Primary School project (Case Study 6) builds on learning from a similar team that delivered St Luke’s Primary School (Case Study 5). Finally the design approach for Greenfields House (Case Study 2) draws heavily on a design strategy established by the Charities Aid Foundation building, which was reviewed as one of the original PROBE Studies in the 1990s.
This structured learning between projects allows architecture practices to develop their knowledge, and as a consequence offer a valuable service to their clients. Evaluation and learning between projects is an important factor for many of the architecture practices represented in this book.
Conclusions
BPE and POE must become critical processes for clients, architecture practices and the industry if we are to improve what we do and solve the societal challenges we currently face. This does not necessarily mean complex and expensive design, but rather focusing on an essential understanding of what we are trying to achieve and how to test it on occupation.
A common theme in this book is the complicating effect of design for ‘statutory compliance’ (Building Regulations) rather than designing for end users. What seems to us simple can be unfathomable to the end user in operation, and in an age of seemingly constant technical advancement we may confuse ‘functionality’ with improved ‘usability’. Although Bill Bordass’s maxim to ‘keep it simple and do it well’ should be recited before every design team meeting, we are still learning what simple really means.
01 Internal view: evaluating brightness and glare