Enhanced Problem Solving Approach to Tackle Repetitive System Failures

The paper describes an innovative approach to performance improvement using Causal Learning (CL), a method based on the general observation that a business performance is largely the outcome of the organization, processes and procedures, ways of working, constraints and norms – the systems that the business applies to itself. These system causes are often remote from physical causes of equipment failures and as such remain hidden until revealed by appropriate analysis. The objective of CL is discovering these system causes that ultimately lead to an undesired outcome or event. CL helps us "learn" the performance system, develop insights from these discoveries and recognize the specific aspects of a system that require change to shift business performance. The Company adopted this approach to improve problem solving and root cause analysis of machinery failures. The initial decision to apply CL followed several outages of power generation systems that continued to occur after previous analyses of similar events in the past. An Enhanced Problem Solving Team (EPST) was established and trained to apply Causal Learning principles to reveal the underlying system causes of these outages. In the time since that first analysis the tools and techniques of CL have been applied to other undesired or unexpected business outcomes including HSE and project work with little or no direct technical content. CL reveals the contribution of well-intended human behaviours behind unwanted outcomes (e.g. hardware failures), and importantly the underlying system causes of these human behaviours. This is predicated on the basis that people do their best to achieve the goals they believe they need to achieve. When it is revealed why those goal were important and why the actions taken were "their best" with the time, tools, processes available to the individual at that time at that place the systems causes can be properly understood. These findings often surprise the organization, particularly when it is made visible to Leaders at all levels how they created, or are responsible for, the systems that influenced those human behaviours.