Distributed decision making

The Cynefin approach to distributed decision-making builds on two established bodies of work. Firstly the use of ‘Entangled Trios’ as a mechanism for problem resolution and the creation of informal networks and; secondly the use of distributed micro-narrative for lessons learning, monitoring and situational assessment. There are additional links to a body of work in peace and reconciliation and conflict resolution. The approaches were initially outlined in the EU Field Guide to managing in complexity (and crisis) and have also seen practice for over a decade in managing patient journeys and more recently in understanding attitudes for the Royal College of Nurses as well as a range of other health sector projects including pioneering work on Residential Care for the Dutch Government (publication referenced below). We emphasise the distinction between distributed decision-making and delegated decision-making.

Name and history

Most rules for decision-making are based on experience and study of past cases and said rules are, in the main drawn up on the assumption that future events will follow a similar pattern within the constraints of a Gaussian, or normal distribution. These approaches which grew out of business process engineering and then six-sigma tended to assume that the exceptions would be just that, exceptions that could be limited or eliminated by good planning and control or even ignored.

This is true for many ordered situations, where the number of interactions and the level of sustainable constraint allows for predictability and the destination of outcome (and output) based targets and control. However, when the number of interactions and agents increases, the buffer stock of material and human resources is exhausted and the issue becomes complex; in the sense of complex adaptive systems. In these cases exceptions increasingly become the rule and the distribution of incidents starts to follow a Pareto distribution where past practice can inform, but not mandate a response to current needs. Under these circumstances higher levels of adaptability and agility are required.

The world can easily become very messy, and the role of leadership and governance structures is to make that mess coherent. With multiple agencies & actors at play a form of coherent heterogeneity will be more appropriate. The concept of coherent heterogeneity, a key aspect of complexity science, is that the system can be very different in one context, but very similar in another. As an illustration, think of the rivalry of club sides in sports that disappear when the players are in their national team.

The design of complex systems involves a few relatively simple objects and processes to interact over time, so that the application emerges from those interactions reflecting real use. As actions stabilise over time they are the focus of further automation and codification. This object-orientated approach to design has been used in system development since the 1980s and is now being extended to include human and other non-digital actors in the system.

Workflow

The elements of the system using entangled trios and distributed ethnography are as follows:

1. Roles and Trio's: Roles within agencies and functions are defined and individuals are allocated to one role or multiple roles. Roles may be qualification-based or process-based and we can also add experience levels in context. Role combinations are called "Trio's" and are defined based on agreed-upon rules that take into account the wider social ecology, different perspectives and their relative awareness and influence.

2. Problem areas: Problem areas are identified and individual problem areas are distributed to role combinations where the Trio are empowered to make decisions within designated constraints

3. Ongoing monitoring: SenseMaker® Genba can be deployed with multiple capture options (observation, decision, recommendation, micro-concern etc.) that are self-interpreted by the active agents at the point of capture. Data is real time and quantitative in nature, but the numbers are backed up by explanatory (qualitative) narratives. In order to reduce burden, we would recommend that this type of capture be ritualised, to replace some conventional reporting.

The experiences of the Trio in the build-up to the decision can also be recorded as micro-narratives on the SenseMaker® Genba app (or can be incorporated via open APIs into an operational system).

4. Rules for decisions: Some decisions can have delays on implementation built into them; they can be flagged for review by senior decision makers who can intervene, using the evidence collected and associated with the decision through the Genba app. If no intervention permission is de facto given unless the rules state otherwise. All of the data from all of the observations is available for real-time monitoring and analysis.

References

Experience Matters: Narrative learning, development, and accountability in nursing home care

Blog posts

Distributed decision making – flagging a new area of development.