11th Oct 2016
Can Productivity be modelled presents findings of some recent research on being able to quantify and model productivity. We share some initial results of the modelling touched on in this blog. However, a more detailed description of our work, including how to model health, wellbeing and productivity concepts in the VE, and the full modelling results can be accessed by completing this short form.
Health, Wellbeing and…. Productivity! Time has come where we can take a closer, empirical and quantifiable look at productivity. Our recent blog and event on Health & Wellbeing (How to do WELL with IES) has generated significant interest and participation from a wide range of stakeholders. Similarly, the Health and Wellbeing movement, including the WELL standard have been gaining momentum and popularity with building owners, operators and designers. But what is it all about? Investing in the health and wellbeing of our buildings and occupants is often seen as a means to an end. That end is Productivity. From service based organisations who want their office based staff to be more productive to retail stores wishing shoppers to spend a little bit more money, it’s time to start taking Productivity seriously. Integrated Environmental Solutions (IES) have begun to explore this concept further by asking a very simple question: Can Productivity be modelled?
Productivity is a complex issue. Take a simple example of an office based organisation filled with hundreds of employees with workstations, working away in a standard 5-day week. An organisation has a means of measuring its own productivity in various ways:
A recent World Green Building Council report on ‘Health, Wellbeing and Productivity in Offices – the next chapter in green building’ highlighted the important costs associated with office based organisations. Staff salaries account for close to 90% of the inputs into a business with rental and property management costs at just under 10%. Utility costs make up 1% of the overall costs.
The diagram illustrates the relative importance of a 10% change to each of those categories. For example, a 10% saving on energy cost, although important, it only represents one tenth of a percent change in overall organisational costs. Contrast this with a 10% improvement in staff related costs and you’ll get a very large 9% difference in the overall business bottom line.
An example of the monetary cost of this, the same study states that annual absenteeism rates in the USA are 3% per employee in the private sector costing $2,074 per employee per year, this value is marginally higher in the public sector. The business case for investing in productivity therefore, is real and tangible. Having the ability to measure and then model it would present a competitive advantage which can spearhead and drive justifiable and verifiable investments to optimise productivity and shape larger sustainability and HR strategies.
Productivity itself can de dependent on many factors both physical and psychological. Physical factors can be further split into physiological and environmental. A persons’ physiological disposition including general levels of health, fitness, weight and mental health, all contribute towards their ability to work and be productive.
Environmental factors including office temperature and thermal comfort, indoor air quality, ventilation rate, levels of daylight, sky view and more, contribute significantly to overall workspace productivity. Interestingly, they are also variables which are easily monitored and measured in modern buildings directly through sensors and Building Management Systems (BMS). These same factors can be modelled and designed, using advanced building performance analysis software tools like the Virtual Environment (VE) which provide a single platform to model the energy, thermal, daylight, comfort, air and climate considerations during a building’s design.
Physiological factors are easily measured through staff surveys, as are factors such as employee turnover, absenteeism and sickness. This means we have a rich body of data developing which can help us to better understand Productivity and assist in developing a sound modelling capability for it.
Recent research has shed light on the impact of physical environmental variables on productivity. Some academic peer reviewed journal articles in particular have been successful in articulating this link.
One such example is ‘A study on the effects of thermal, luminous, and acoustic environments on indoor environmental comfort in offices’ Li Huang et. al, Building and Environment V49 2012. This study provides a good overview on the impacts of thermal, daylight and acoustic impacts on overall indoor environmental comfort, specifically in offices. Another study; ‘Investigating the productivity of office workers to quantify the effectiveness of climate change adaptation measures’ T Kershaw, D Lash,. Building and Environment V69 2013, talks more explicitly about office based worker productivity and what adaptation measures suitable for addressing the impacts of climate change would have. These two are just a couple of examples of the wide ranging and increasing numbers of studies being conducted worldwide on health, wellbeing and productivity. A tremendous example of the level of activity globally is summarised excellently in the review paper ‘Occupant productivity and office indoor environment quality: A review of the literature’ Y Al Horr et. al. Building and Environment V105 2016. This very recent paper reviews over 300 separate studies from 67 journals and conference articles and focusses on the physical environmental variables affecting office based productivity.
The WGBC report referred to earlier also cites many academic and non-academic works on the same subject. It states, ‘a comprehensive body of research can be drawn on to suggest that productivity improvements of 8-11% are not uncommon as a result of better air quality’, meaning low concentrations of CO2 and pollutants and high rates of ventilation. Such high gains in productivity would certainly justify the investment surely? A similar case is made for improvements in thermal comfort and daylighting and lighting.
Perhaps the most interesting and bold attempts to quantify productivity based on physical environmental variables appear in two recent studies. In my own research I came across this recent 2016 article; ‘Integrated optimisation of energy costs and occupant’s productivity in commercial buildings’ H Akbari, Energy and Building V129 2016. I recognised the author, Professor Hashem Akbari, from Concordia University, whom I had first met back in 2010 at the Passive and Low Energy Cooling Conference in Rhodes. I remember his passionate presentation on launching the Global Cool Cities Alliance. As I read his paper further I realised an incredibly simple yet powerful model had been presented. It referenced another study ‘Some quantitative relations between indoor environmental quality and work performance or health’ O. Seppanen, W,J Fisk, HVAC&R V12(4) 2013. The latter study had presented the results of 26 other studies and used a meta-analysis of the data to deduce a quantified expression for productivity based on thermal comfort (Temperature) and indoor air quality (Ventilation rate) for office based environments. Professor Akbari’s excellent paper had used this to run optimisation algorithms for minimising energy costs whilst maintaining or improving productivity in commercial buildings. Both papers looked too good to be true. As a modeller, I decided to test the hypothesis, from where we began this story; can productivity be modelled
The papers relate Relative Productivity RP(%) as a function of inside temperature (⁰C) and ventilation rate (L/s/person). Figure 1 shows the relationship of relative productivity against inside temperature and Figure 2 shows relative productivity against ventilation rate. Using regression techniques both these relationships have been transformed into usable equations.
The relative productivity against temperature equation is a simple cubic relationship. It states (T is temperature):
Relative Productivity (%) (Temperature) = 0.165 * T – 0.00582 * T² + 0.0000623 * T³ – 0.468. (1)
Because it only contains one variable (Temperature) it is easy to model.
The relative productivity against ventilation equation is again simple (Q is Ventilation rate):
Relative Productivity (%) (Ventilation) = 0.02 * ln(Q) + 0.960 (2)
The effect of temperature and ventilation on productivity is given by combining the RP(%) of the two individual equations.
Overall RP(%) = (Average (RP(%)Temperature + RP(%)Ventilation) + Max (RP(%)Temperature, RP(%)Ventilation) ) / 2 (3)
Thus we have been given a straightforward method of predicting productivity based on two very common environmental variables, temperature and ventilation rate. Both well understood, easily measured and easily modelled. A note of caution, the authors of the respective papers highlight that these equations are only indicators of productivity based on the available data. Nevertheless, they do present us with a good starting point. Ideally these equations will evolve over time and explicitly include more physical variables backed up by empirical data.
So how can we combine the power of a whole building simulation tool like the VE, with these empirical relationships linking productivity to common simulation outputs?
A thoroughly empowering feature of the VE, is Custom Variables, the ability to use any of the hundreds of variables in the VE and construct one’s own using mathematical and logical operators. (https://www.youtube.com/watch?v=iq4fRq8Qnl0). In fact, there is almost no limit now to the kind of metrics you can define and report out of the VE.
I have created my own custom variables to report on ‘productivity’ using the equations above. Take Equation 1 on temperature as an example. In Vista-Pro, I have used custom variables to create a new variable called ‘Productivity thermal comfort’. I have inputted Equation 1 into the expressions tab, seen in the screen shot below. From the variables list, I have selected Air Temperature denoted by the letter A.
The variable ‘Productivity thermal comfort’ now appears in the list of variables in Vista-Pro (Circled in red in the image below). The VE has been used to run a simulation on a three storey office building using the London (DSY) climate file with standard occupancy and internal gain profiles for office based occupancy. The new variable can now be plotted, analysed and graphed as any other variable would while analysing simulation results.
The chart below shows Productivity thermal comfort (%) – blue line plotted against office air temperature (⁰C) – red line, in hourly values from 1st – 31st July.
Productivity is a percentage value between 0-1. Some observations of the relationship can be made:
Imagine being able to model productivity in a straightforward manner. It would provide a useful KPI to track during the multiple design stage scenarios which are normally considered. Given the environmental variables it is based on, it would also be simple to monitor and verify post occupancy.
IES are experts in modelling real world performance of buildings and its environment and occupants. Encouraged by recent developments in modelling productivity, IES are interested in further data which can help us refine the productivity equation. To find out how we can help you understand health, wellbeing and productivity better for your organisation, get in touch to hear more thought leadership in this exciting and developing field.
IES has already helped our clients achieve the WELL Standard Credit 54 (Circadian Lighting Design), Credit 62 (Daylight Modelling) and Credit 76 (Thermal Comfort) across half a dozen projects. If you’re an Architect, Consultant, Contractor or building owner, contact me to find out how IES can assist you in enabling and embedding good practice Health, Wellbeing and Productivity concepts in your projects and buildings.