Dr. Hodkiewicz–of The University of Western Australia–and team asked 176 maintainers about their work and what makes procedures effective.
Gridium: Hello everyone, and welcome to this conversation with Professor Melinda Hodkiewicz of the School of Mechanical and Chemical Engineering at the University of Western Australia, located in lovely Crawley, near Perth, alongside the Indian Ocean.
My name is Millen and I’m with Gridium. Buildings use our software to fine-tune operations.
Melinda and I will be discussing a recent paper she co-authored titled, Are You Sure You Want Me to Follow This? – A Study of Procedure Management, User Perceptions and Compliance Behavior.
I had barely finished reading your paper, Melinda, when I reached out to invite you to join the podcast. Really, thank you for joining us today.
Dr. Hodkiewicz: Oh, it’s a great pleasure. Thank you.
Gridium: Melinda, let’s start with this question: how did you get interested in maintenance?
Melinda : Ah, well that’s a long answer, but essentially when I finished university I went to work in the mining industry.
And in the early 90s, I found myself as a maintenance engineer at a very large gold mine in remote northeastern Nevada, a place called Elko, Nevada. So, it’s regarded as a truck stop between New York and San Francisco. But, it’s a wonderful place, one of the world’s biggest gold operations. And one of my roles as maintenance engineer was to set up a machine condition monitoring program.
And condition monitoring uses lots of really fabulous, high-tech equipment to listen to machines to try and see if they’re healthy or not; we listen and look at them in unusual ways such as with thermography, oil analysis, vibration.
I had all of this very clever equipment, but I really had to work very closely with the mechanics and the electricians, machinists and fitters who were part of the team to try and identify and troubleshoot equipment to make sure that we could identify if it had problems, where they were and fix them before they failed.
Gridium: Right.
Melinda : And in the process, I learnt a huge amount about how machines work or don’t work from maintenance guys that were part of the team.
Gridium: And why do you think maintenance matters?
Melinda : AMaintenance–as the name implies–is really about the repair of equipment. It’s retaining its value. The equipmenthas a function, that’s why we bought it, and how do we retain that function?
Many people think of maintenance as a cost. “Oh, damn. I have to repair, this is going to cost me.” But they don’t think about the value part, which is, “If I maintain it, and I maintain it well, I then get a function of this asset”–whatever that function is.
It could be propeller produced by a turbine or the functionality of a robot or a train running on time. Without maintenance, those things don’t happen; assets don’t function. And so, maintenance is a really crucial part of business for an asset owner and it is value-adding.
If you do it well, the asset will function as intended and if you do it poorly, it won’t. So, it’s both value-adding and value-destroying. And maintainers have a handle in both of those things depending on how they do the work and how the work is set up for them.
Gridium: You’re paper–the beginning of the title “Are You Sure You Want Me to Follow This?”–how did this get off the ground?
Melinda : This got off the ground because I’ve long been interested in doing a study like this; however, what you need to do is,as an academic, you need funding for work like this and to provide that funding, you need to find company groups that will fund that.
We located a report in 2012 produced by the Department of Mines and Petroleum in Perth, in Western Australia that looked at 12 years of fatalities in the mining and petroleum sector. And one part of that report indicated that in 89% of fatalities there was either no procedure in place, or the procedure and rules were not followed at that time.
That was really a motivating factor too–some quantitative evidence that said procedure following is an issue. This was also assisted by the fact that the majority of the people who were killed in these accidents were in fact maintainers. And so, this provided a mechanism to go and talk to a particularly leading mining company–leading in the sense that they are very progressive around safety–and say, “We would really like to look at this. Would you support us to do that?”
And they agreed and they came in as both a partner and an enabler of this project, in a very proactive way. They didn’t have to do it, they hadn’t had any particular issues, but they could see that this was interesting data and they also understood enough about the background of what we were doing to support us to do this work.
I’d like to acknowledge my co-authors on this study: Lisette Kanse, Katharine Parkes, Xiaowen Hu and Mark Griffin. They’re all colleagues from the University of Western Australia–though Kathy is from Oxford–and we really brought a lot of different skills together to the development of the model, the conducting of the interviews, and all the statistical analysis afterwards. And it was really a team effort.
Gridium: That’s great. And you cite many procedure studies that are not focused on maintenance: why do you think that’s happened?
Melinda : Ah, yes. Well, it’s the issue of the invisibility of maintenance.
As you talk to people like Lee Vinsel and his colleague Andy, and anybody to do with the Maintainers II group: one of the things we talk about is how invisible maintainers are in the current society.
I think that goes in an academic sense as well. It’s very attractive and eye-catching to do studies of surgeons or airline pilots or nurses; these seem a higher tech, more sophisticated perhaps–a challenging group to work with. Whereas maintainers are largely invisible both in society, I think, and almost certainly in the academic literature; there’s really very, very little in the academic literature about maintainers and who they are–in the engineering academic literature–and how they work.
Gridium: So, let’s talk a little bit more about the paper itself, Melinda. Can you summarize the hypothesis you and your team were seeking to test?
Melinda : Yes. We started off with the idea that what we were interested in was procedure following behavior; so, this is: compliance with the procedure or non-compliance, as two separate constructs.
Then we said, well what about the procedures might affect compliance or non-compliance; what attributes do they have? Do we have a perception of them being positive or negative?
And then the antecedents of that. The procedures aren’t objective and seen in isolation; they exist as part of the socio-technical construct of the operation. How people are involved or not in the procedure; and how management responds positively or negatively–what we call a punitive or a learning approach to non-compliance–may also affect people’s perceptions of procedures and their willingness or not to comply.
So essentially, we looked at three things: management of procedures and management response, procedural attributes (whether they were received positively or negatively), and the collective effect of those things on compliance or non-compliance.
Gridium: What is an example of a learning-oriented management response?
Melinda : So, a learning-oriented management response would be when in the event that you did not comply, the supervisor for instance would ask in a very open way, “Ah, I see that you didn’t…” for example, “…change the oil in truck A yesterday and the procedure said to change the oil. Can you just explain to me why we didn’t do that?”
And you might say, “Well, actually, I heard from the guy on night shift before that they had to bring the truck in unexpectedly; there was a problem with the filter and he changed the oil. So, I haven’t done it again.”
So, instead of that understanding that both respectfully–initially, the respect for the maintainer that they had to follow the procedure, that there is… the default position being “there’s probably a good reason for that”.
Gridium: Right.
Melinda : As opposed to the default being “gosh, you’re just trying to cut corners all the time”. So, like… a learning is around “let’s understand why you didn’t do it and let’s talk about it.” rather than the default being, “I just assumed you have poor reasons for doing this.”
Gridium: Your team interviewed and surveyed 176 maintenance professionals. Can you summarize those two different maintenance teams? You know, their industry and the research process?
Melinda : Yes, I can. So, we worked with a major mining company and we worked essentially with two sites. Both of these sites are remote sites, thousands of kilometers away from the nearest city.
One site was primarily responsible for the maintenance and repair of water, communications, power and other services and the maintainers were predominantly electronics and communication technicians, electricians.
And then the other site was an operating mine, and they were predominantly heavy mobile diesel fitters and electricians working around the very large 300 tonne haul trucks. The sites–in the interview with the maintainers, which I maybe will talk about a bit in a minute…
Gridium: Sure.
Melinda : …it was very evident that the atmosphere on the site was really quite different. So, although you had the situation that you had the same company, you really got quite a different sense of a difference in work cases and a difference in a number of factors around sites.
That was interesting to us as well as the main focus of the study. So, what we did is we–as far as method goes–we developed an initial survey based on our model, and then we travelled to the sites and we conducted a number of semi-structured interviews with the maintenance teams to trial the survey both to understand… to flesh out various aspects, but also to make sure that what we were asking was in a language that made sense to them so that our questions would be interpreted properly, as well as to build support for actually answering the survey.
Gridium: Right.
Melinda : Then, once we had that done, the survey was administered more broadly to all of the teams on both sites, as part of safety training days, so it was done in work time. Everything was done with ethics approval and we were very clear to everybody that participation in either the interviews or the surveys was entirely voluntary.
Gridium: These 176 maintenance professionals were employees after all; how did you get the involvement of the employers? Can you talk a little bit about that?
Melinda : I think I can’t underestimate how important the support of the employer was in this particular study.
And this support had to cover both very senior people at a global level, to get permission to do the work; at a national level, within Australia; and also, at a site level to allow us access to site.
I think the, obviously, the crucial fact was some very strong leadership support at the national and international level to do this work. And a conviction that there could be something that could be learnt about the issues we were talking about. But, in a good way, not an expectation that we would get any specific results.
So, I should say that I think one of the challenges with doing work, this sort of study on maintainers in the past has been the fact that maintainers are primarily employees, and in order to get access to them as employees and understand their work in the context of their workplace, you need the support of the employers.
And that support is challenging… it’s challenging to persuade any organization to let academics come in, in any situation, and interview their people under ethics where everybody… where people can say what they want and their anonymity is guaranteed.
And that can be very challenging and I can’t say how grateful I am for the support and farsightedness of the company that sponsored this: that they trusted us to do this and that they allowed us and supported us to get up on site. They paid for travel, they did a huge amount of logistics to get us on site on a number of occasions. They allowed us access to people… you know, all of these things are very difficult. And unless that happens, this sort of study doesn’t.
Gridium: Melinda, you and your team conducted what looked like in the paper to be some really interesting interviews. What were those about?
Melinda : So, the… what would happen in the interview, is we were a number of people volunteered and were identified as being available to be interviewed. We were given some quiet, secure rooms on the sites to interview people. They came in, we explained that…I did them with a colleague who also had a background in maintenance-and I’ll come to why that’s important in a minute.
They had the opportunity to… we explained to them about ethics approval. We spent some time talking to them about the process and assuring them around anonymity and confidentiality of whatever it is they were going to say. We usually start… well, we always started with some discussion around “So, tell us about your job. What do you do? Tell us about your daily tasks and responsibilities? What do you think is important about your job? What is demanding or difficult about your work?”
And then onto some more specific questions around procedures, “How do you use them? How do you access them? Are there any issues with this? What do you do if a procedure doesn’t seem appropriate for the job?” Those more general questions around procedures.
But, I think the main thing is that we spent quite a lot of time at the beginning of the interview making the maintainer feel valued, trusted, comfortable and respected.
Gridium: I had never considered the difference between good procedures and good procedural management. Can you explain?
Melinda : So, procedures don’t exist in isolation; there is actually no objective valuation for procedure.
You can’t say it’s objectively good or it’s objectively bad. What we have are perceptions of those procedures: whether they are accurate, whether they are logical, whether they’re at the right technical level, whether they’re user-friendly. Or, are they out-of-date, unnecessary, time-consuming.
What we think of them will be grounded in who we are, what we’re using them for, how experienced we are… and, a variety of factors like that. We might also feel positively or negatively about them depending on whether we were involved in their development or not. You know, if we have some buy-in, we’re bound to feel better in things.
So, understanding that we bring all of this baggage with us when we use a procedure is, I think, really important.
Gridium: I know that you made a reference to… well, is there a link between the baggage you referenced and culture. What role, if any, do you think culture might play?
Melinda : I find culture a very difficult thing to explain. I suspect that if I try and define it, it might well sound different to what you have in your mind or indeed what our listeners think of as culture.
So, I have spent a little bit of time around psychologists–organizational psychologists–which, of course is quite a difficult thing for an engineer. And I’ve heard culture described by them as something that’s pervasive…it’s all around us. It’s very difficult to define, yet we share it.
But I think most of all, we express it through our activities. I like to tell my engineering students that culture is what we do when nobody else is looking. So, it’s very difficult to unpick the cultural aspects of this in a way that is fairer and objective so, of course, the things we’re talking about are very influenced by the culture of the place.
As engineers, we like to think that people are going to be completely predictable, kind of like computers; but of course, they’re not. And as I mentioned early, we… it was very obvious to us that there were very different cultures on these two sides. But it’s very difficult to objectively say exactly what they were.
Gridium: Your paper does a good job of laying this out, but I–since we’re talking about it–need to ask… what did you learn from this study?
Melinda : Well, we learnt that… we had some really interesting things actually.
We learnt that if you have a positive vperception of procedures you are very much more likely to comply with them. So, the question is, how do we ensure that people do have this positive perception.
And here we find that there are two very strong relationships: one, between being involved in procedure management–so that would be around being involved in the development of procedure, being involved in the update of procedures, know where to find them… all of those things.
And then management’s positive learning approach to procedural compliance. So, in fact, as you do those procedures, actually being very positive around how you’re doing them and reinforcing the good as well as not reinforcing the bad.
So, having a learning approach towards non-compliance and a learning approach towards compliance, is really important. So that was… which I think points the way to management in thinking about new procedures and how they manage procedures, to recognize that positive behavior-in both of those things-is self-reinforcing.
Gridium: I see.
Melinda : The second pathway is slightly more complicated to explain and it suggests that if you have a negative perception of procedures that that’s associated with the lack of a learning approach. So, that’s not quite the same as a punitive approach, if that makes any sense.
Gridium: Yes.
Melinda : So, if you don’t… the absence of a learning approach produces a negative perception of procedures. And a negative perception of procedures is really, really highly associated with non-compliance.
Gridium: I see. Some of the interviews were quite honest and direct. Did any of them surprise you?
Melinda : No, they didn’t.
I think part of that honesty is the fact that we did take a lot of time to create a safe space for people to talk to us. And when you create that safe space, people feel that they can be honest–sometimes maybe brutally honest with you.
I’ve found that maintainers, particularly good maintainers, understand and value their skills and their abilities. And when asked, can be quite articulate about those and about the context in which they perceive their skills as not being valued.
That’s not to say I find them in any way egotistical; in fact, the very best of those maintainers tend to be sort of very quiet and unassuming, but clear and firm if that makes sense. So, they did have some very interesting things to say and we were very glad that they said them.
Gridium: There was an example about a red pen. I was struck by that example.
Melinda : Yea, there were probably a couple of quotes that are worth talking about: one of them is the red pen quote.
And basically, it says that if a procedure is incorrect, the expectation is that they, the maintainers, will red pen the procedure and hand it back to the document writer and have it amended. That amendment part is where it gets stuck.
The maintainer can sit there and red pen it and hand the document in. And then all he sees the next time the procedure comes out, is that nothing has changed. So, he says, “Why do I bother?” And that’s when they switch off. And there is another somewhat similar to that that essentially says maintainers are two dimensional: it either gets changed and they see a direct result, or it doesn’t and they switch off.
So, you know, I think that there is a sense–as with anybody, not just maintainers–that if you put effort into trying to get something changed and management doesn’t respond, that that’s a really demotivating thing and situation.
And that you have that in maintenance; I think it’s exacerbated in maintenance by the fact that often the people who are doing the changes are not really familiar with the process. They can’t do it themselves, and so… in some way, that makes the situation worse. You’re not dealing with people who really understand what you’re doing; you’re dealing with technical writers or engineers who actually are not deeply familiar with the procedure or the work you do or the context in which you do it.
Gridium: Imagine, Melinda, that you’re responsible for developing new maintenance procedures for a complex technical system at a new organization. Any advice?
Melinda : Well, I think the advice would be to take some of the lessons that we have from this paper, which is to identify a core group of maintainers who are respected and valued by their peers in the workshop to really get them involved in the procedure development process; and not just the writing or the use of the procedures, but how they use them and the structure around that… how they’re updated.
Making sure that there are feedback mechanisms. Making sure that there are… that supervisors have the right support and training to create the environment around more of the learning culture with their teams. Some supervisors have this and some don’t.
So, I should mention that maintenance supervision must be one of the hardest jobs in the world. You are dealing with people who often have quite large teams, sometimes physically spread in a number of locations often doing work that requires safety permits that require review and signatures.
In an operation environment that can be quite rushed, it’s complex technically; but they’re also trying to find parts, deal with people issues, get operations on side. It’s a completely challenging role, and on top of all that we want them to be respectful, knowledgeable; people who can find time to talk and understand, rather than just jumping in and assuming something is not right. So, the maintenance supervisors are a key cog in this particular wheel.
And we actually interviewed… have a whole bunch of data both from the maintenance supervisors from their teams and from the teams about their supervisors that we haven’t really delved into yet; but, we think that there are probably some very useful insights from that. But, one thing we do know is that the team dynamics is a really vital element in an effective maintenance operation.
Gridium: Do you think industry is getting better at maintenance?
Melinda : So, do I think industry is getting better at maintenance?
I would like to think that a day in the life of the maintainer is different than how it was when I was in the maintenance job thirty years ago, but I struggle to identify how it’s different.
If we separate what maintainers do from maintenance in general, it’s changing, maintenance,but the actual day-to-day work, I don’t see as changing nearly as much as I think it should have.
If I look at what’s happened to operators in my thirty years… when I first started, I was in operations and we ran around turning equipment on and off with creative red buttons and turning dials to change the speed. Within a couple of years of me doing that, we had a control room where we did everything from a computer and now, we largely do it from a control room in the mining industry that’s 2,000 kilometers away with a fraction of the people that we used to do it with.
So, the role of the operator has changed out of sight in my thirty years; but the role of the maintainer and what he-and it still largely is–does on a day-to-day basis hasn’t changed much.
All of the work we do is condition monitoring and whatever that is supposed to help us predict failures earlier and take more proactive action, but actually getting that to happen on a regular basis predictably and reliably hasn’t come along, I think, as much as it maybe should’ve done in the last thirty years.
But I’m hopeful! I work in that area now and I’m hopeful that we shall start to see some changes and that those changes will make a meaningful difference in the work that maintainers do every day.
Gridium: What are you working on now?
Melinda : Ah, so probably most…I’ve got two things I’m working on: the most interesting thing that-maybe from our podcast listeners’ point-of-view-is the work that I’m doing with Bonita Carroll and an Associate Professor, Martin Forsey.
Bonita is an anthropology PhD student and she has just spent the last several months with maintainers at a mining… and now at an oil and gas facility in northwestern Australia. So, she’s been up there with maintenance crews. She’s been with shutdown crews, she’s been with night shift crews, condition monitoring crews… in the workshop talking, listening, interviewing, observing maintainers. It’s a really in-depth study aimed at a number of things.
But one of the things is to get a good understanding of what the lived experience of maintainers in heavy industry is today, and she’s had the opportunity to go and see it in the mining sector and also in the oil and gas sector. And so, we might see some interesting differences or same things… same situations around that–that could be interesting.
Her PhD is also focused on female maintainers, which are a fairly rare breed. And… but in order to understand the female maintainers, she has to understand the maintenance context first. And so, there’s a lot of interest and excitement around what might come out of this.
So, if we better understand the lived experience of maintainers, we can understand motivations, recruitment, retention… it could inform a lot of policy decisions; training that could materially impact on maintainers of the future. And the project is actually called The Maintainer of the Future project.
That’s one thing I’m working on. And the second is I run something called the UWA System Health Lab and that really is around developing the next generation of technologies to help predict equipment failure. And with that I work with statisticians, material scientists, electrical engineers and we do all sorts of nerdy, techy, fun things around new technology.
Gridium: Yes, that sounds interesting and of course, I’m also really much looking forward to Bonita Carol’s work on the future of maintainers.
Melinda, it has been a pleasure speaking with you today. I want to thank you for your time and this has been quite interesting. Thank you.
Melinda : Well, thank you very much Millen. Thank you very much.
