Self Organising Networks: the valuation of collaborative teams for complex infrastructure projects through Social Networks analysis presented by Stephen Pryke of UCL and Simon Addyman of Self organising Networks.
Stephen opened the session with the observation that digital information is fundamentally changing how humans connect and communicate. It is now possible to track the connections that people make with each other when working on a project using a system known as Social Network Analysis (SNA). The concept is derived from a blend of mathematics and social research. The typical organogram of a large project is often merely a hierarchal analyses rather than representing the real flow of information. The reality of a project is complex interconnections.
The SNA process requires collecting data from the participants and then mapping out who has actually shared information with other people involved at particular stages in the project. The process is made possible by the development of software which can analyse and then map out, large amounts of data that can then be displayed in a format that can be interrogated.
Stephen and Simon demonstrated the concept using two projects for Transport for London, these being the refurbishment of Bank underground station ticket hall, and remedial works to Hammersmith Flyover. This was a two-year research project sponsored by Innovate UK. An initial pilot study was carried out, with a follow up exercise six months later.
Social Network Analysis is a science that inevitably has its own terminology. A social network consists of a finite set of actors and the relations between them. An actor is a role holding entity, who could be an individual, firm or even some inanimate object. The objective is to identify and understand the prominence of particular actors, and understand who has a high number of connections and therefore has a position of formal or informal power in the network at a particular point in the project cycle.
A visual map is produced by the software, which helps identify who the prominent individuals are. Software identifies the communities including those that were densely connected and those that were sparsely connected. It creates the ability to drill down into the various groups and sub groups in the project. Initially, the work identified three clusters, Design, Delivery and Decision making.
The researchers started with the structure as set out in the traditional hierarchical organisation chart. Using an electronic survey process, they then asked everyone involved in the project to feed back who they had sent information to, or received information from, over the previous week. The process was particularly looking for weak connections that would not typically show up on a traditional organisation chart. There was an initial resistance to providing this information, but once the participants understood the output, they became fascinated to see how their networks actually worked in real time, and also how they were evolving.
Why is the Social Network Analysis process useful?
The value of SNA is that it allows managers to understand the flows in communication which are slowing the progress of the project, how information is processed and how decisions are being made. Since the construction process fundamentally relies on the ability of humans to connect with each other efficiently, the information that social network analysis provides is potentially very useful. Examples identified included;
- The ability to check that the patterns of communication were appropriate for any particular stage in the project. For example it was possible to see whether the Project Manager, the Contractor and the Client were connecting on an even basis.
- Identify the ‘brokers’ who helped information flow, and also the ‘blockers’ who create bottlenecks by holding information back.
- Identify the ‘influencers’ who were connected to a wide range of others in the network.
- Identified people who were not on the formal organisation chart, but were important to the project.
- Identify the ‘isolates’ who are not talking to others. For example, the map highlighted situations where people sitting next to each other were not sending information to each other when it was assumed they were connected.
- Identify who reported to who, which is useful in matrix organisations where management responsibility can often be split. The process can also show up disconnections and where anticipated reporting lines are not in place.
- Allowed a ‘collaboration check’ to see whether the right people were talking to each other.
Stephen and Simon made the observation that many high performers intuitively understand their networks. This tool offers the potential to helps those of us who lack the natural intuition to build networks in an effective way way. In time, we anticipate that models will emerge which identify the optimal network, making Social Network Analysis a proactive process.
The potential applications are many and varied. At a high level, the data provides a strategic overview as to what is really happening on the project, providing a richer quality of information to support the project leadership. It can also help the teams ‘on the ground’, for example as an induction tool for someone joining the project who needs to understand with whom he or she should be connected. A final observation that came from the floor was that this new capability might create a shift in the way that large complex projects are run, with less reliance on the project manager, and the emergence of a new lead player – The ‘Network Manager’.
We will watch future developments with interest.
This insight has the potential to be hugely valuable as a project tool in the future. We are very excited to be working with CPC and UCL to offer innovative solutions to ever more complex project environments.”