Dr Shrikant B Sharma, group director of Smart Space at BuroHappold Engineering, explains the latest innovations in crowd dynamics and modelling and the importance of full integration with BIM.
Efforts to predict people movement and behaviour in the built environment have existed for almost 50 years, what is the latest thinking?
Sophisticated crowd modelling techniques have been in use for about 20 years, but even now the focus is generally on assessing whether spaces will function safely at peak times or during an evacuation. The day-to-day operations that define the fundamental success or failure of a building through its lifetime are largely left ignored.
We are now looking increasingly at comfort- and experience-driven outcomes and modelling the effects of a whole range of conditions on human behaviours, such as temperatures, noise levels, oxygen levels and opportunities for physical interaction. This is happening much earlier in design than with conventional crowd modelling.
Companies like ours have developed tools for use during early stages of design, working with architects to measure experience and inform strategic decision making.
What are the benefits of modelling comfort and experience?
In a workplace you can optimise the layout, lighting and heating conditions to ensure that people are more productive, comfortable and interactive, which ensures that the building owner or occupier of a building can get maximum value out of the asset.
It’s also possible to model spaces to maximise collaboration and chances for interaction between people, which is actually the opposite of congestion-driven models where you are trying to minimise chances of people bumping into each other. It’s possible to introduce measures such as strategic partitioning of desk layouts and coffee machines, or locations of entrance and elevator lobbies to extend interaction times.
We and others in industry are beginning to tap into data and sensors in buildings so that we can start to measure comfort and interactions. In theatres and museums you can mine user sentiment patterns from social media data or bluetooth sensors that will show dwell times to get a more holistic model of behaviours and experiences.
The benefits of modelling comfort and experience apply to every sector – enhancing footfall in retail, productivity at workplaces, repeat visits at museums and theatres, operational efficiency and revenue at airports and rail stations etc.
The Internet of Things and big data are creating a huge buzz – are they feeding into crowd modelling systems?
This trend has emerged over the past couple of years. The ability to access live sensor data, and social media streams enable us to see which parts of a city become active at different times and what people are discussing and engaging with. Sophisticated natural language processing and sentiment analysis allows us to see which spaces make people happy or unhappy, excited or bored, and why.
Big data analysis gives us a better picture of what makes things tick, from city and site scale, right down to specific buildings.
At Buro Happold, we understand that future proofing and delivering the maximum value for a client, means making designs more data rich, evidence driven and outcome-based. That means exploiting all available data out there.
We have a partnership with a major corporation to pull data from IoT sensors in their buildings to record the temperature, light levels, humidity, CO2 levels and anonymised location data for different times of day.
The data is input into our SmartMove software platform which is able to run various interactive visualisations, with predictive analytics in the background that tell you which layout will maximise interaction, improve spatial utilisation, comfort and so on.
Another project at a hospital involved following the movements of nurses carrying anonymous trackable cards they pick up in the morning. This allowed us to examine their walking patterns, how much time they spend with patients, navigating corridors, or getting to medical cupboards etc.
This data gave us invaluable actionable insights and the crucial information needed to optimise internal layouts and improve staff efficiency and interactions.
Does the software interact with BIM?
Yes, SmartMove runs inside a BIM environment, you can run a crowd simulation right on top of it, it automatically recognises 3D objects, such as stairways or a door and reacts accordingly. A key aim has always been to integrate with BIM completely so that people don’t have to import and export files between platforms and lose important data in the process.
We are also working with Flux.io, a Google-X spin-off company, to develop a spatial analysis app to run on top of their platform, designed to connect various BIM models, such as Revit, Rhino and Sketchup with Google Maps.
The app runs a streetmap-like environment, accessed online, that enables users to drag and draw over any area in the world and see how well connected buildings and streets are in an area, how far someone will have to walk to get from A to B to C, or even run environmental or mobility analysis.
The idea is a client or other non-technical person can manipulate the 3D model, dig in and examine the data, without having to download BIM software or other analysis tools.
What’s the next big thing on the horizon?
Real-time simulations that work within a sketching environment, integrate with live sensor data, and provide predictive insight with real-time interactions and ‘what-if’ scenario testing.
Crowd modelling software should not require effort, you should be able to draw a little sketch in a BIM model or a Google Map-based environment and it should be able to tell you if it is in the optimal orientation, with optimal accessibility, improved interactions, increased footfall etc. Every bit of detail you add should trigger new insights.
We understand that future proofing and delivering the maximum value for a client, means making designs more data rich, evidence driven and outcome-based. That means exploiting all available data out there.– Dr Shrikant B Sharma, BuroHappold Engineering
