Root Cause Analysis

Management Insights to Root Cause Analysis

Knowledge Management

A technology whose time has come

 "Successful knowledge management treats knowledge as a resource by exercising selectivity, imposing order on information resources, adding structure to information to increase its value, and proactively capturing information that might be useful in the future."   

         PriceWaterhouseCoopers "Technology Forecast: 2002-2004"

 Can there be anything more important to an organization than the knowledge of what internally threatens its survival and success, and what can be done to control and prevent those threats? The concept of having corporate knowledge available at our fingertips to expand our perceptions and support our decisions is certainly a seductive one. Even more appealing is the thought that somehow this knowledge would give us wisdom. Who can resist having the right information at the right time and having the wisdom to make the right decision? This is the promise of root cause analysis knowledge management technology. The issue rests at the core of operations quality and the future of continuous improvement activity.

 Knowledge management seeks to place what we know about our organization into an integrated resource for decision-making. Data that have intrinsic value of their own, gain value through integration with other forms of data. New relationships are revealed to provide new perceptions for better decision-making. Knowledge management doctrine assumes that ordered data is knowledge, and not just addressable, indexed facts. In most cases, that is a daunting assumption. Today’s efforts to harness the potential of knowledge management still rely primarily upon integrating existing records into an addressable structure that can provide elevated perceptions and visibility. It is a logical first step, and in analogy the lowest hanging fruit on the knowledge management tree. One can anticipate the evolution of knowledge management going from this first phase to a systems environment in which new forms of data will be introduced, and new data methodologies will be applied both to produce a richer content, and to provide the means for better mining of the data resources. It is this avenue to new capability that ordered root cause analysis data provide.

Little understood is the fact that data can itself generate a form of wisdom. If we strip away the quaint human characteristics associated with the concept of wisdom and focus just upon the specific capabilities we associate with wisdom, we find that data through form and interaction can produce knowledge that goes beyond the mere awareness of separate facts, and that indeed through the accumulation of knowledge and the perception of patterns within that knowledge can come transferable wisdom. There is of course a reason why we tend to associate wisdom with grey hair, reflective patience and being at peace with the world.  Historically it has taken a person a life time to gather facts, experience situations, make mistakes, and direct curiosity and intellect. But now computer knowledge management technology promises to jump start our wisdom by providing data parameters that can substitute for those long years of trial and error experiences, gleaning perceptions from the written words of sages and even intellectual exercise. It may not be appropriate to extend this analogy between computer data base technology and insightful savvy beyond a few decision making parameters, but perhaps inference of further similarities might not be too far a field. In any case, the ability to judge, predict and compare the appropriateness of actions, to be able to accurately anticipate outcomes and the ability to transfer the functional relevance of principles to dissimilar systems have served to define and depict wisdom in man’s philosophies and theologies . These are the emerging capabilities issuing from knowledge management.

 Management’s Role

 The manager has a critical role in the evolutionary process of knowledge management. It is management’s role to set the direction, provide the ground rules, and to allocate the resources. In order to succeed in their role, managers must take charge of the process. It is a role that requires understanding of concepts and current technologies.  The goal of a knowledge management system is to provide the substance for continuously improving management decisions for planning, strategies, and goals. It requires an ordered and dynamic data resource, not just a storehouse of records. Fundamental criteria for the design of such a system must include stimulus and provision for continuous improvement of the resource information itself. Without management’s firm lead, the early goal of integrating all available existing data can itself become a fortress against change. It means that designers must accept the difficult challenge of a moving target in order to produce a system that will succeed in the long term.  It means that management must keep its finger in the design pie in order to assure that criteria include the ingredients for growth of the knowledge management system itself.

 The Quest for New Knowledge

 At DSI, the search for more meaningful information has long been an overriding priority. From DSI’s research recently came just the kind of advance that knowledge management seeks in order to expand the utility of data. For decades our clients have been applying REASON’s logic rules to guide the data gathering and analysis of problems. The REASON logic ordered the data into a verifiably accurate model of the problem. We reasoned that because each client was using the same objective standard procedure, all of the cases produced by the different clients should share data qualities in common. It was this premise that generated the research study of a large body of REASON models to uncover those common qualities. From the study came the discovery that systemic causes associated with the operations problems from many different industries and processes were combining into distinct causal patterns and in predictable sequences that were shared among the many different sources. We immediately perceived in application that when a person knew these patterns, he could predict what types of causes combined to produce a particular pattern. With that ability he could then know exactly the right questions to ask in order to uncover those causes when gathering information about an operations problem. This is a classic example of how existing data can be mined to gain new insights and knowledge. From the new knowledge came perception of new avenues to pursue and new practical applications that promise to be pathways to future advancements. 

Our systems designers took all of the knowledge we had gained from over twenty years of field research and development, combined it with the new knowledge of causal patterns and ran with it. With the strict criterion for growth that we recognize as essential to knowledge management, we have developed the design of a seamless data system for operations improvement’s new horizon. The expert system software teams interactively with the user to discover systemic causes, watching input and responding with the important questions based upon the users input.

Focus upon systemic root causes

The system starts with a guided process for gathering and qualifying data about an operations problem. Based upon the recently discovered causal patterns, an expert system predicts what kinds of causes must combine to produce the problem, and asks the essential questions to accurately identify those causes.

 Integrated into the data gathering process is the on-going validation of data. Data are ordered into sets of causes that combined to produce each step in the process that led to the problem. At each step, the individual sets of identified causes are tested for accuracy and completeness. This causal logic test assures that critical elements of the problem cannot fall through the cracks, as well as assuring that irrelevant facts are not included in the analysis process.

 The information gathering process itself, automatically orders the data into a unique cause and effect model that supports objective quantification. Each avenue is pursued until either a root cause is identified, or the factor is found to be not correctable internally by the organization, or information is insufficient to continue the avenue of investigation. In effect, the process separates information much the same as do our minds naturally when we are solving our daily problems, into things we can do something about, things we cannot do anything about, and things for which we have no answers.

 Once a root cause has been identified, the seamless process provides a systematic method for establishing where in the organization prevention can best be accomplished. This pinpoints exactly both what can be done in the action plan to gain immediate control over the internal system that produced the problem, and what steps are necessary to sustain control in the future. Because REASON focuses upon the internal business processes as the seat of root causes and their removal, those steps necessarily involve developing or modifying business processes. The resulting data are ordered and unified into causally associated facts that translate into new knowledge about why problems are occurring in the organization. It is this knowledge that sets the REASON Lessons Learned System apart.

 When the data gathering and modeling of the problem are completed, the system applies an objective quantification standard to measure exactly how much each cause contributed to the problem. The system automatically generates a logic tree model showing graphically how each cause combined and led to the occurrence of the problem, a plain-English narrative explaining step-by-step how the problem happened, and an analysis comparing all prevention options in order to identify the best and most cost-effective solutions. The process has in effect mapped the information into a causally associated sequence of causally associated sets of causes. The knowledge that is contained within the data structure goes beyond content to reveal the causal sequences, combinations and interaction of the effects of the business processes themselves, revealing a new level and dimension of understanding exactly what is happening within an organization, and providing a view of options for addressing and improving the organization’s internal systems.

 At this point in the decision making process, the principles of organizational control, the objective quantification of the systemic model and the rigor of analysis combine to provide a path toward wisdom. With the data, one can reliably predict the outcome of actions, can compare options for anticipated best control and cost-effectiveness, can assess the criticality of risk remaining within the environment and can detect similar patterns within seemingly dissimilar environments and processes. Little graying of hair will be noted, but significant wisdom for decision making is provided.

 The completion of the prevention action plan is recorded and the data file is automatically submitted to REASON Lessons Learned. Key personnel throughout the organization have previously submitted a data profile of all of the issues that are important to their responsibilities and processes. As the REASON root cause analysis data flow into the lessons learned system, REASON is monitoring content, watching for matches to those data profiles. When a match is found, it immediately sends an alert. In moments the targeted individual has received a heads-up alert. Information gets to the person who needs it and can act to head off a developing problem. This real-time picture of what is happening within the organization is also immediately addressable as an information resource for workers, supervisors, engineers, planners and managers. Instead of a simple indexing of words to support word searches, REASON data is an ordered structure of data that permits you to search for factors that combined into a causal system at a point in time to produce a step toward a problem. The results of these searches are richer data content and strongly targeted information. Thus the organization’s information about counter-quality becomes a dynamic, real-time resource, rather than merely a static storehouse of facts.

 Elements of Risk

 The same data file that was created when the problem was first solved, and that was used by Lessons Learned for access to past experiences, also serves as the resource for new kinds of data for risk managers. The simple and straightforward premise is that each root cause represents a lack of control or risk within the environment. Because the REASON system can measure the system that produced the problem, it is possible to calculate how much each cause contributed to the problem. In effect, it measures the amount of actual risk each cause represents in the environment. If we act upon a particular root cause, we remove a measured amount of risk. Knowing which causes remain in the environment enables us to measure and manage the known remaining risk. The hard numbers returned by the system provide unique and meaningful analyses and risk trending capabilities.

Corrective Action Management

 One important characteristic of knowledge is the transference phenomenon that occurs when numbers are associated with perceptions. Invariably the numbers gain value and utility as application is expanded into other areas of activity. For example, the ability to measure risk and to assign priority to corrective actions based upon that measurement provides an objective criterion for assigning responsibilities, setting completion due dates, tracking and managing corrective action plans. In the REASON Corrective Action Management System, where root causes are identified and prioritized, corrective action plans written, responsibilities assigned, activity scheduled, aged, tracked and monitored, the utility of having objective criteria for establishing priority and scheduling of work projects is a significant benefit.

Control Effectiveness

 REASON root cause analysis provides new and unique data for oversight and monitoring of control effectiveness at the micro and macro level. For example, it is possible to measure the effectiveness of control not only by facility, process, and operation, but also at the organizational level to determine effectiveness of the overall control at the management, supervisor and individual level. This view is available for a single problem, a group of problems, or all recorded problems that were solved and recorded. Again, the data can help form decisions for planning and action. With REASON data one can pinpoint areas of concern, identify systemic patterns, determine best actions and monitor in real time the effects of those actions upon controls. Some would say that the exercise of such capability is a demonstration of management wisdom.

 When problems and their solutions are linked to business processes, the data can provide dramatic insights to the internal health of the organization. As organizations, we set survival and success goals; we devise strategies to achieve those goals; we develop policies to accomplish those strategies; and we set in procedures to comply with those policies. By directly linking the occurrence and amelioration of operations problems with these fundamental structural steps, we create a window through which we can see how an operations problem on the shop floor directly impacts that chain of executive controls from policy to strategy to corporate goals. We see exactly how policies are affecting our strategies and whether our strategies are contributing to or detracting from our goals. The ability to associate business processes with the strategies and goals of the organization provides executive managers with both a new visibility of the efficiency and effectiveness of executive controls over critical operations, and a new means of monitoring the function of organization strategies and goals.

 New Avenues to Improvement

 This expanded resource of information made possible through DSI research and product development has early provided new capabilities and insights into ways organizations can improve operations quality. We anticipate that as the discipline of knowledge management matures, the resource of this validated and quantified data about internal systems will become increasingly valuable to managers seeking better control and quality of operations.

 If your management group is seeking new avenues to continued success and operations improvement, we encourage you to investigate the promise of REASON as a means to gain increased value from your internal information. If your organization has already made a commitment to knowledge management as a growth activity for operations improvement, we invite you to explore the dramatic benefits that REASON Root Cause Analysis holds for you.

 For more information on how REASON Root Cause Analysis can be deployed to upgrade your operations improvement activities, please call (903) 236 9973.

Copyright 2005 © DECISION systems, inc. All Rights Reserved