Uncertainty and risk ‘Heat Maps’

Risk heat maps

A risk ‘heat map’ shows possible impact against likelihood of various events or scenarios, as in this one from the EIU website:

The ‘Managing Uncertainty’ blog draws attention to it and raises some interesting issues. Importantly, it notes that it includes events with both positive and negative potential impacts. But I go further and note that in assigning a small blob to each event, it fails to show  Knightian uncertainty at all.

Incorporating uncertainty

Uncertainty can be shown by having multiple blobs per event, perhaps smearing them into a region. One way to set the blobs is to get multiple stakeholders to mark their own assessments. My experience in crisis management and security is that:

  • Stakeholders will tend to judge impact for their own organisations. This can be helpful, but often one will want them to also assess the impact on ‘the big picture’ and the ‘route’ through which that impact may take effect. This can help flesh out the scenario. For example, perhaps an organisation doesn’t see any (direct) impact on them but another organisation sees that although they will be effected they can shift the burden to the unsuspecting first organisation.
  • Often, the risk comes from a lack of preparation, which often comes from a lack of anticipation. Thus the situation is highly reflexive. One can use the heat map to show a range of outcomes from ‘taken by surprise’ to ‘fully prepared’.
  • One generally needs some sort of role-playing ‘game’ backed by good analysis before stakeholders can make reasonable appreciations of the impact on ‘the whole’.
  • It is often helpful for stakeholders to mark the range of positions assumed within their organisations.
  • A suitably marked heat map can be used to facilitate debate and scenarios and marks developed until one either has convergence or a clear idea of why convergence is lacking.
  • The various scenarios will often need some analysis to bring out the key relationships (‘e.g. contagion’), which can then be validated by further debate / gaming.
  • Out of the debate, supported by the heat map with rationalised scenarios, comes a view about which issues need to be communicated better or more widely, so that all organisations appreciate the relative importance of their uncertainties, and how they are affected by and affect others’.
  • Any difficulties in above (such as irreconcilable views, or questions that cannot be answered) lead to requirements for further research, debate, etc.
  • When time is pressing a ‘bold decision’ may need to substitute for thorough analysis. But there is then a danger that the residual risks become ‘unspeakable’. The quality of the debate, to avoid this and other kinds of groupthink, can thus be critical.


The UK at first misunderstood the nature of the protestors in the ‘first fuel crisis’ of 2000, which could have had dire consequences. It proved key that the risk heat map showed not only the mainstream view, but also credible alternatives. This is seen to be a case where Internet, mobile phone and social media changed the nature of protest. With this in mind the EIU’s event 2 (technology leading to rapid political and economic change) have positive or negative consequences, depending on how well governments respond. It may be that democratic governments believe that they can respond to rapid change, but it ought still be flagged up as a risk.  

 See also

Cynefin , mathematics of uncertainty

Dave Marsay 


About Dave Marsay
Mathematician with an interest in 'good' reasoning.

One Response to Uncertainty and risk ‘Heat Maps’

  1. Pingback: Science advice and the management of risk | djmarsay

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: