April 01, 2008
Southwest Airlines Continued and The Problem with Wrench Time
Comments on Southwest Airlines and the beginning of the topic on revisiting our premises related to R&M from a business point of view. Editorial by Howard W Penrose, Ph.D., CMRP
Last week we discussed some of the issues related to maintenance cutbacks by Southwest Airlines. While they had claimed that redundant inspections would have caught problems and no one was in danger, 22.7% of the aircraft had either cracks or fractures that the mandated inspections were supposed to catch. The impression is that no one is in danger until they are killed and the redundant inspections did not catch the problems or were not effective to detect the issue.
Since then, while the news has been fairly spotty concerning the issue, investigations of both the FAA and Southwest have been initiated. The FAA has also made the decision to audit this particular inspection (and possibly more). The curious point is that a number of airlines have announced a reduction of flights and retiring of aircraft with the public announcement that it has to do with fuel costs. Could it be that this ‘shot across the bow’ with the Southwest incident suddenly has the airlines interested in their maintenance practices? If so (and I will be watching this), then we were lucky that the decision is based upon the stock market (Southwest’s share price dropped significantly) and not blood. That would be a pleasant change! (see Editorial Part 2 for the rest of the story).
In the meantime, we are going to start reviewing some of the premises that make up our reliability belief system. I am going to present some concepts that go against these systems with appropriate arguments. I will also introduce some business strategy concepts that support those arguments.
The first one we will address is ‘wrench time.’ There is a push to increase the wrench time in many organizations by focusing on pushing more work through the maintenance organization. However, as outlined in ‘The Goal’ by Elayahu Goldratt, the concept of overloading the maintenance organization (ie: planning and scheduling too much work) by assigning more hours than are available is the same as carrying too much inventory. The piles of work end up piling behind each technician and the opposite intent results, an increase in backlogged maintenance and no improvement in the elusive efficiency measurement of wrench time.
As identified in the business concept of the Theory of Constraints, focusing on ‘efficiencies’ is counter-productive. In effect, many of the metrics that are being pushed, such as wrench-time, are actually hurting our ability to do work rather than helping.
The outline of the Theory of Constraints is to:
1. Identify the System’s Constraints – what are the limits to ‘throughput’ of the maintenance organization;
2. Decide How to Exploit the System’s Constraints – how to adjust the non-constraints of the system to supply the constraints;
3. Subordinate Everything Else to the Above Decision;
4. Elevate the System’s Constraints – increase capacity through the constraint; and,
5. If in the Previous Steps a Constraint Has Been Broken, Go Back to Step 1.
In the short form, exploit and then expand upon the capacity of the resources that you have available.
An easy way to identify your constraints, which can be human, process or other, as shown in ‘The Goal,’ is to look for where the largest amount of ‘inventory’ piled up in front of the constraint. For instance, if we follow the concept that is often espoused of scheduling up to 400% (yes, this is actually recommended by a popular book) of the capacity of the maintenance organization in order to meet an efficiency goal (increased wrench-time), the result will be the opposite of what you expect. Instead, you will end up with a lower and lower completion rate of PM’s and critical equipment will be missed.
As outlined in ‘Physical Asset Management for the Executive,’ we discuss how to approach planning and scheduling around the constraints, including training and reactive maintenance, of your maintenance organization. This is important as the automation of maintenance scheduling, such as the use of CMMS/ERP systems will only automate and accelerate the problem of maintenance completion issues.
There are additional methods of exploiting and elevating the constraints of your program. However, when performed correctly, proper planning and scheduling can reduce workloads, increase maintenance throughput, while giving maintenance personnel the satisfaction of work completion and a job well done.
Let us consider both approaches with the following information. We have two infrared thermographers who are also qualified to remove and install electric motors. In an uninterrupted week, for their plant, each technician can complete and report on 100 panels in 25 hours of work each during a 40-hour workweek (wrench-time metric of 62.5% - not too bad at all!). If a motor fails, up to 6 hours of the 25 hours available is taken up. The company requires a 90% completion rate of PM’s.
Using the popular approach, up to 400 scans per technician would be set up, in case they are not all available and to ‘push the technician to work more effectively.’ With the limits, after one week, assuming no other emergencies or issues, 300 scans are not complete and are added to the following week’s workload per technician! A 25% completion rate of PM’s. If one motor fails, then the completion rate falls to 76 machines, or a completion rate of 19% and 324 scans are moved to the next week.
What if this continues and the backlog becomes more than unmanageable and the technicians are focused on more and more reactive work related to incomplete PM’s? That’s right, the concept was eliminated from production over a century ago! You cannot dictate an improvement of efficiency by overloading a constraint. Instead, you bog it down and the effectiveness of that constraint, in this case trained technicians, until it is completely ineffective and has a negative impact on the company’s profitability. The production side of our companies understand that increasing inventory (excessive PM’s) means carrying costs and other expenses that directly, and severely, impact the bottom line. The result of an ineffective maintenance organization and starts the thought process by management of bringing in outside organizations.
Now, if we look and understand the constraints of the organization, in this case no more than 100 scans per week, and the possibility of a drop of 24 scans per motor replacement, we can plan around the issue. Remember, the cost impact of PM performance is not the cost of the maintenance technician, but the impact of the PM on equipment availability. By focusing on labor cost, you lose sight of maintenance throughput and the impact is negative. By focusing on maintenance throughput you end up with more effective use of maintenance resources and can begin to evaluate methods of elevating or breaking the constraint.
Therefore, we consider the chance of a reactive motor replacement per week and decrease the completion requirement by those hours. For instance, if we know that we have an average of one motor failure every two weeks, we schedule 87 scans per week per technician. For weeks where no reactive work occurs, work on the following week’s schedule, or catching up on the past schedule can be performed. Work-studies can be performed to find ways to improve maintenance task throughput and may include anything from a hybrid outsourcing program, training additional personnel, correcting PM routes, scheduling and coordination improvements, etc.
In effect, lazy management approaches of overloading personnel to get them to ‘do more,’ promotes poor and unsafe work practices, directly impacts company profitability, and can result in other negative impacts on the organization. A more proper approach, by considering the relationship between maintenance methods and production methods, improves the effectiveness of the maintenance organization, improves equipment availability and much more.
Next week we will address the next problem with present maintenance processes.
References: “The Goal” by Eliyahu Goldratt; and, “Physical Asset Management for the Executive” by Howard W Penrose, Ph.D., CMRP.
Sincerely,
President, SUCCESS by DESIGN
Executive Director, Institute of Electrical Motor Diagnostics, Inc.
Member: National Writers Union (UAW 1981) and International Federation of Journalists
Author: “Physical Asset Management for the Executive (Caution: Don’t Read This If You Are On An Airplane)” and “Electrical Motor Diagnostics: 2nd Edition”
