Over the last few years I have been working on a project at Jisc, called The Intelligent Campus. Though I left the project in March, in my new role and as part of Jisc’s work on Education 4.0, I still have an interest in that space and what is possible and what benefits it could bring universities and colleges.

Last year I wrote an article for Educational Technology on what the difference is between a smart campus and an intelligent campus.

At the recent Westminster Higher Education Forum on the 11th December I delivered a short presentation entitled Latest trends in intelligent campus design.

I wanted to talk about how the campus of the past is evolving into the smart campus of today, but look forward to a future where we may have an intelligent campus.

The smart campus is already here; the technology, sensors and data analysis capability are all available, but it isn’t all joined up and so has limited scope in terms of what we can learn and how we can use the knowledge. What we need now is the intelligent campus, where data from the physical, digital and online environments can be combined and analysed, opening up vast possibilities for more effective use of learning and non-learning spaces.

I first asked the question, what is a campus? Across the UK (and the rest of the world) there is no standard model for an university campus. Some university campuses are based on a single campus, others have multiple sites, some across a city or a town. Some campuses are just the university, others have more than the university, they have neighbours. These may be related businesses and organisations, but sometimes it will be just other neighbours, some retail, sometimes business, sometimes other educational providers, and sometimes just people.

Even on a single site campus you may have multiple buildings. Some will be new, some not so new and some really old.

When some of these were built they were really “dumb” buildings. They would have had heating and lighting, but it would have been designed at that point with very few if any controls for the users of that building. Well there may have been light switches!

You could argue that so called smart buildings have been around for a while now, a thermostat is something of a smart technology, they enable the heating (or the air-conditioning) to come on when the temperature in the spaces reaches a desired setting. Though in some places the users of the spaces can adjust that setting, sometimes the settings are fixed and can not be changed. In a similar way sensors enabled lights to be turned off if no moment could be detected, though sensitivity was challenging, as is often seeing with people in classrooms and offices waving their arms about when the lights go out.

In many ways a smart building is one in which certain aspects of the building are noted and recorded. Building a data picture of the building, from use of electricity, heating, water, lighting, energy usage. Some newer buildings can also measure utilisation, occupancy, flows of people.

That data could be stored in a data hub and then portrayed in a dashboard of some kind, visualising the story that the data is telling us. Using that visualisation, we might better understand what interventions to make to reduce energy costs, improve utilisation and so on.

One of the challenges for the smart campus, is that it is unlikely that every building has the same capabilities when it comes to the collection, storage and visualisation of data. Partly because buildings were probably built at different times, but the lifespan of buildings is much longer than the lifespan of smart building software. The end result is that with larger campuses, you may have multiple sensors, multiple data hubs and vastly different dashboard and visualisations.

You could argue that the first actual smart estate is one where the building data is aggregated and the dashboard and visualisation gives an overall view of the campus, but you can drill down to individual buildings, maybe even individual rooms. You may need to think about how to record that data in a standard consistent format, so that it can be aggregated and used.

So that’s a smart campus?

What would make it intelligent?

Well a campus doesn’t exist in isolation. As well as all that buildings and estate data there is a wealth of other data been collected as well.

The challenge often though is that data is in silos. Locked away in it’s own data hub and dashboards. If we could allow the data to break free…. Could we create a campus with a deeper enriched set of data about lots more than just the buildings data. Could we add in data about the users of the campus, how they are using the campus, where they are using the spaces on the campus. Transport utilisation and congestion. Weather and other environmental factors, such as air quality, could be taken into consideration. What is happening that day on the campus, will there be more or less people than usual. Will there be guests around? What about differentiated energy costs? Are there some days when energy is cheaper? What about the curriculum, who is in that day for teaching, who will be in the library studying? Are there any examinations? You could go on.

A smarter campus will take data from a range of sources, not just the physical aspects of the campus and how they are being used, but also the data from digital systems such as attendance records, the virtual learning environment, the library, student records, electronic point-of-sale and online services.

This joined-up approach can provide insights and intelligence into the student experience that we would otherwise miss. These insights can inform and support decision-making by individuals across the campus, including students, academic and professional service staff. By using live and dynamic data, decisions can be made that are based on the current state of the campus.

But it isn’t just about collecting data, could we create personalised dashboards for different parts of the university, adding in campus and estate data into their datasets to create a deeper richer picture of what is happening. Allowing for positive interventions that take into account a wider dataset rather than a narrow data set.

This would enable us to build a smarter campus.

So what do we mean by the intelligent campus? 

Well a deeper and wider dataset provides us with a better picture of what is happening on a campus, it provides us with the intelligence to make date-informed decisions about changes we want to make to the campus, the student experience, and so much more.

However could we start to imagine a future where we can start to predict how a campus will be used, so rather than react to what is happening campus, we can be proactive and make interventions before something becomes a problem. Using a range of processes and techniques to improve that overall student experience.

The next level of the intelligent campus is to move beyond using this live data to add a machine learning element to the analysis, providing further insights. Bringing in historical data, feedback and evaluative data, we can start to add a level of predictive analytics that will allow students and staff to make informed decisions on what is likely to happen.

However….

It’s not always going to be so simple is it?

Some of our data is locked away in proprietary systems, we are unable to access that data and add it to a central data hub. Should be even be buying systems today that don’t allow us to access the data within that system?

Data and visualisations are not something that everyone knows what to do with. You can provide dashboards to people, but will they use them? Will they know how to use them?

Ethical use of data is critical, these systems can stalk, sorry track, individuals across a campus, they can see where they are, what they’re doing and who they’re with. There are some real privacy concerns, that need to be addressed.

Legal use of data is also important, you can’t afford to ignore GDPR and other legal requirements in the use of individual data..

Having said all this, there are some real technical issues that need to be resolved. How do you define a space? How do you define utilisation and occupancy? How do you know what people are doing and how they are using a space.

Another aspect is validity, just because there is correlation, this doesn’t mean there is a relationship between the data sets. The data only tells you part of the story.

Universities across the UK are building smart campuses, however the real cutting edge stuff is when we start to move from the smart campus, bring in more data, more analytics and more predictions, and start to build the intelligent campus.