Every business activity has a potential to fail. After initial design, process error reduction or correction measures are, for the most part, ignored. The effort to correct a problem begins in response to a flaw that is uncovered during performance of the workflow or system. It is unfortunate that only then do companies consider how to prevent the same error from being repeated. Why is failure analysis not given the attention it deserves? Most supervisors or managers are too occupied with how to make the system work and fail to consider potential pitfalls. Individuals and process sites have become accustomed to reacting, trying to fix errors or create workarounds after the problems have occurred. These fixes often require significant process changes or software redesign that can be costly. Wouldn't it be more efficient and cost-effective to anticipate errors before they happen and proactively institute preventive measures? Many organizations fail to go that one step further and perform a Failure Modes Effect Analysis (FMEA) on their workflows.
A FMEA provides answers to these basic questions. How can a process or product fail? What will be the effect on the rest of the process or system if such failure occurs? What action is necessary to prevent the failure? In other words, it facilitates an analysis of the system or process, before an error actually takes place, to identify areas of potential failure and gauge the impact. This proactive process can be employed not only to examine the maintenance of equipment, but also in the design of new services and procedures that may affect a workflow. The best time to employ a FMEA is prior to deployment of the new work processes, so that a preemptive action can be taken. Even if your organization is not prepared to conduct a full FMEA, you can derive value from a simplified approach.
A SIMPLE FMEA PROCESS
For a FMEA to be effective, a team approach is essential, so start by building a process review team. The effectiveness of the review and the process depends upon the expertise of the team members, the quality of the team output and on the willingness of each team member to give his or her best effort. Be selective. Teams may include:
System Safety Engineer
Materials & Process Engineer
Others as required
Then you need to perform five steps on the process flow to assess the potential for an error. In this example we will use work order creation.
1: Create a Diagram of the process. Develop a work flow diagram or chart of the process. Do not rely on memory alone to review the process flow; use a flow chart that indicates all activities taking place at each operation. This diagram should represent all major components or process steps as individual blocks connected together by lines that indicate how the steps are related. Make sure to identify all individual pieces of the intended process that need to be reviewed. Ask yourself: "Is all work recorded? What information would be important to collect in each step? How would the information be defined, stored, and used, from work order creation through end of life? What is done differently for an emergency? Who will create the work order? What information would need to be given to the maintenance technician or the operator? How would the work order records be administered?" These are the functions of that activity or step.
2: Identify Failure Modes and Causes. The potential failure modes (that is, how and where the systems or processes may fail and what can go wrong) need to be identified while considering how the work order would be used. Questions you might ask at this point may include: "How do we ensure parts get recorded? If an equipment name looks or sounds like another equipment name, could the techs work on the wrong one? How do we make sure the right person approves or works on it? How do we ensure work is done in the required timeframe? How is failure information recorded? How is work time tracked?" At this point the failure mode should be identified whether or not the failure is likely to occur. The potential causes for each failure mode should be identified and documented.
3: Define the Effects. For each failure mode, the team should determine the likelihood of an error, and the potential consequences. First, ask how often it could happen, then: "What would happen if the wrong equipment was worked on at the wrong time, or by the wrong person? What would happen if failure information is not recorded properly? What would happen if parts are recorded against the wrong equipment or not recorded at all?"
4: Rate Consequences. Team members should consider the severity of the outcome and identify preexisting controls that could help eliminate or detect the error before work completion. The team should discuss and record the failure effects, failure causes, and any current methods of controlling each potential failure mode. For example, would obtaining additional information or using computer alerts, bar coding, or a double-check process catch any of these errors every time? Numerical values are assigned to determine the likelihood of an occurrence, its severity, and the chance that it would be detected before causing a problem. These numbers are multiplied together to rate the failure consequence.
5: Identify Actions. If failure modes reveal errors with significant consequences, actions need to be taken to: prevent the error from occurring, detect it before the work order is released, or minimize its consequences. Conduct brainstorming sessions to develop effective and innovative ways to reduce the failure and identify corrective action to improve the process. Such actions might include: improved communication methods, drop down lists, automated workflows for distribution and escalation, or bar coding.
The value of a work process FMEA is that it forces you, up front, to perform a thorough analysis before trying to code software or train everyone on a new procedure. Although industries have developed elaborate FMEA scoring systems to rank items for action, a simplified FMEA process as described above can be an efficient, proactive risk management tool.