Episode 14: Driving operational performance excellence across diverse fields, analyzing changeover downtime the right way, and dissolving departmental silos.

Season 1 Episode 14, Jul 28, 2021


Prashant Dhillon, Performance Excellence Manager at Frederick National Lab for Cancer Research, talks about applying performance excellence strategies across diverse fields such as manufacturing, construction and healthcare, how we should be looking at changeover management, and how companies can encourage cross-department communication.

Topics in this podcast


Show Transcript

Prashant: The first step of Changeover Management, is to observe and analyze which of those steps are internal. We have to stop the machine. And external is something which while the machine is running, we can still sort of do it.

Intro:[Background music] : Pashi presents The Means of Production, a podcast about what it really takes to build, maintain, and scale the processes that produce the physical products that power our world. Every episode, we ask a manufacturing expert to walk us through the nuts and bolts of how they do their job. We explore how and why they got into manufacturing. Dive deep into the hardest problems they've solved on production lines and discuss their thoughts on what's broken in manufacturing today and how those things can be fixed. This podcast is hosted by me Siddhit Sanghavi, Pashi's US manufacturing operations lead, and former assembly engineer at Ford Motor Company. If you are a part of the manufacturing world and you're interested in being a guest on The Means of Production, get in touch with me at siddhit@pashi.com.

Siddhit: Welcome to season one, episode 14 of the means of production. And with me today is a very, very good friend, former batch mate at Virginia Tech and performance excellence manager at Frederick National Laboratory for cancer research, Prashant Dhillon. How are you Prashant?

Prashant: Good evening. I'm doing well. Hello Sid.

Siddhit: Prashant I'm really excited that you could make it. I know you're going through, like a really busy time and we're going to talk about all that, but before I do that Prashant, can you go ahead and read out that disclaimer please?

Prashant: Sure. So I work at Frederick National Lab for cancer research. But this is my own opinion and it's not the opinion of a Frederick National Lab, what we will talk about in this podcast. I'm not a spokesperson for this organization official or otherwise.

Siddhit: Perfect.Thanks Prashant. So firstly, can you just tell me and the audience, how you're doing, what is going on with you and all the fantastic things that you're up to these days and there are some great news to be shared there?

Prashant: Absolutely. So life is really hectic right now. So we have a new addition to our family. We have our second daughter. Our first daughter is now six years old. So technically we have forgotten everything about raising a child. So lots of sleepless nights and you know how the game goes.

Siddhit: Congratulations Prashant.

Prashant: Thank you.

Siddhit: And what's going on on the career front and workfront and what are you doing in that area?

Prashant: So on the Workfront actually at Frederick National Lab, we are setting up this new department or office called the quality and performance excellence office. My responsibility here is to initiate and institutionalized the elements of performance excellence. Right now what we are doing is, we are doing some pilots with the projects andknock on wood. They're going in the right direction. Once we gain that momentum, we would kind of leverage and snowball that effect.

Siddhit: That is very interesting to hear Prashant and you're my first guest who currently works at a research laboratory. So this is very new for me. And I want you to speak a little bit about that in our first question, in which I ask the guests to kind of trace the journey of how they got into this field in the first place, because I know that we were together learning industrial engineering and you've worked in manufacturing environments for. So can you just explain how you ended up here at Frederick National Lab and all the things you did before that and what made you take this whole path?

Prashant: Sure. So technically speaking, there are two phases to my life. One is associated with manufacturing and then the life after that. So let me first try to explain how I got to manufacturing itself. So Sid, as you know, I am from a farming family and growing up at the farm, we are exposed to fluid machinery such as pumps, compressors, etcetera, and also the farming equipment. So somehow this exposure led me to pursue mechanical engineering as my major in college. And while in college, as the first flavor of the corporate world, through internships of course, I got into manufacturing a little bit, for example, I had my first internship with Mazda Garageand then I had the opportunity to co-op at Honda Cars. Later down the road, while at Virginia Tech, when you and I, we were pursuing our industrial and systems engineering, I had another stint with GE Appliances again manufacturing.And then my first job in the United States was with a German manufacturing organization. So that's how I ended up organically in manufacturing itself.

Once I was in this organization, I had the opportunity to kind of trial and test a few roles. I was able to move around within the departments. Soon I realized that I enjoy diversity in projects, and I have a great passion for this cross boundary collaboration. So this led me to the logical step of stepping into the performance excellence role, first in manufacturing and then I wanted to see if whatever I've learned in manufacturing can be applied to other industries and sectors. So I first went into construction and last year I had the opportunity to transition into my current role at Frederick National Lab.

Siddhit: This is quitea journey Prashant. So thank you for sharing and look at looking at your experience. It kind of shows that no matter what the industry is, whether it's manufacturing or it's outside of manufacturing. There are certain types of processes at play, which are quite similar in the way you look at waste and you look at performance and you look at excellence and you've kind of demonstrated that it is possible to apply the same general principles of excellence that you learned from manufacturing and applied them in places like construction and also now in this role in a laboratory. So it's quite interesting, the journey you've taken and good on you to try such very diverse kind of roles. It also gives like people who are just starting their career, a picture of what you would do if you wanted to move out of manufacturing or whatever you want, to try your hand at different things.

I think your career is something they can observe. So thanks for that description. Now you've mentioned that you are a performance excellence manager. I would like to know more about it in terms of the challenges you face. So can you, in the form of the answer for the second question tell us what kind of challenges you faced and how you went about solving them? And this could be not just one problem, but it could be like a series of problems. It could be a tough week. It could be a tough year. It could be something that you faced on a continuous basis at any of your previous roles. So what were these challenges and how did you face them?

Prashant:Sure. So Sid in this particular instance, what I would like to share is, an enriching experience in my professional life when I was back working in manufacturing itself. Now, as I mentioned I started in a German organization where I had the opportunity to experiment and try different roles, I moved around. And towards the end, ended up in the performance excellence deploy, the operational strategy of our plants within our region, to remediate deep plants productivity and lead time issues. Now, here to add to this complexity, was that our plants were not always in United States. Added to the complexity, to the variability of how we can come up with the operational strategy for these different plants, et cetera.

So to overcome this variation, the way our team went about it was to leverage on a standard framework, or you can call it an approach.We call them as modules. So every module had a specific objective in mind, and there was a particular cadence and particular way of how we went about one module after the other, just like any other framework or an approach. For example, after our standard preparation with any of the plants, we would always start with what we call as analysis, where we would assess all the factors that would impact the output of that particular plant, regardless of which industry that plant is catering to. We would look into their inventories, lead time, scrap, turnover, the quality management systems. We would also look into the production infrastructure where the market is going, et cetera. So everything was on the table. We had two things in mind when we were actually doing the analysis, basically what does our baseline look like for everything in place?And secondly, we wanted to kind of dig out the opportunities for that particular plant.

Next, we would always facilitate what we call as sort of a vision, where we would bring about the teams to create an operational strategy based on the analysis that we had already done. And also catering into it was the input from our business unit, research and development, et cetera. So come to think about it, this sort of set our not start or the vision for that particular plant itself, where it would want to go in years to come two to five years. So if you are kind of thinking it through, we have our vision, we have our baseline. Now, all that we have to do is sort of connect the dots, means we have to come up with a deployment plan of how actually we would go from our baseline to where we want to go.

So our next step was always to come up with a deployment plan or third phase or module. Now, technically speaking, it could be two to three years, but we always focused or freezed our first year itself. And for the execution of anybody who is in the Lean or Six-Sigma field listening to me, we would then leverage on what many people call as Action Workouts, Kaizen events, Rapid Improvement Events to kind of go for that execution. And for every workshop, now here is where things were so complex, that based on the problem at hand, the tools would change. For instance, if you were dealing with something to do with changeover, we would leverage med related tools. If it was a simple 5S sort of a solution, we would kind of cater to that.If we were tackling with a development team, sometimes we would do Failure Mode Deficit Analysis, et cetera. So our tools within our belts would change, but the framework of how we actually went about it or the methodology would be very similar to what people refer to as DMAIC or A3 thinking.

Now with this deployment, at the end of the day, you also want to kind of see if we are sustaining ourselves or not, right? So at a certain gains or at a certain point of time, we will always sort of review to see if the approach that we are taking or the execution that we are doing is actually taking us in the right direction, or do we need to kind of change course. So this whole framework was leveraged to kind of give us some stability with this variation of product offering, with the variation of different countries, different plants, et cetera. And then within those frameworks, many of the tools would keep on changing based on whatever is at hand. So this is something which I thought was very much enriching from my own experience, with that diversity of just the culture, the thoughts, and to add it all, the product offerings.

Siddhit: So Prashant that was a really good framework you laid out. Can you expand on maybe a particular application or deployment of that framework, in a specific example, to just understand what happened at zero feet level, at the ground level, just some nuts and bolts of it?

Prashant: Yeah, sure Sid. So before we actually go into the granularity, let's recap of what we have just talked about. So we have on a macro level talked about a framework, which helps us to draw a baseline through analysis. It also tries to identify the opportunities that we can work on. Through vision it helps us identify what our goal is, where the business and where the plant specifically wants to go. And then we also talked about how the deployment plan can be sort of weaved, but where the tires meet the road, is the execution, means that is where things start to take some form. So far we have just talked theoretically or through meetings and through presentations. So I'll walk you through one example, which was what I observed is very common to most of the plants that at least I visited and I'm pretty sure anybody who's from manufacturing, they might have in their career came across such a problem also, which is a typical problem of changeovers.

Many organizations, they will have this as one of the opportunities here we are at X you know, time figure or timestamp. And we want to actually improve by Y. So in this particular case, as I mentioned earlier for the granularity, for the execution, we leveraged the DMAIC, just define measure, analyze, improve, and control. Now the DMAIC is for that specific problem itself. And here, our specific problem is, just as an example, the changeover management. So I will go through the nitty gritties of DMAIC. Many people know about it. I think I'll just kind of give the flavor of the changeover management, that, what were the big steps sort of involved over there.

So when you look at any of the process, let's think about, a good example when we go to either Starbucks or when we look at a pit stop of a Formula 1 racing car. What we observe is that at the pit stop or even at the coffee shop there are many things which are done right at the coffee machine, but then there are many things which are done on the backend.So like that for any of the manufacturing on any of the process, the first step of the changeover management workshop, or we used to call it SMED, is to observe and analyze that, which of those steps are actually internal to the process, or they call it internal elements or external elements, external steps. What that means is, what are the steps or what are the process steps that we are doing when the machine has done internal, we have to stop the machine for it and externally is something which while the machine is running we can still sort of do it. The first basic step is to identify, Hey, what are, from whatever we do, what is it that goes into the internal bucket? What is it that goes into the external bucket? And you'll realize that it's a mix and match, a few of them, that sometimes some steps have internal, then external, internal, external, et cetera, et cetera.

But once you do that, and then comes the discussion part where, the team sits and I recommend video, goes through the video and thinks about, Hey, these were the steps which we were doing when the machine is down, means not productive. Production is not getting done. What components or what elements of that portion can we actually transfer to external? What that means is, say for instance, I am trying to change the ruler of a specific machine. And for that, I walk to say the inventory shop to get that particular ruler and come back and until that the machine is done. So these are the things which we can think about, Hey, this step can be actually converted into or channeled into external. This can be done beforehand, or it can be done by somebody else. We don't have to specifically completely stop the machine for such a long time for this, or at least that stopover can be reduced.So that's the second step. The second step is trying to understand, trying to kind of brainstorm what internal steps or what internal elements we can actually transfer to external.

Now, when you do that, your standard operating procedure also changes. You'll see that there's a bit of shuffling that you have to do. Other questions also start to come into picture because you'll observe that the changeover per say is not just relying on just that operator. It's more of a systems approach, for that particular changeover to be effective, many more departments or teams have to feed into it in a timely manner. And that becomes a bigger problem, means that becomes a bigger coordination, not a problem, but an opportunity. So once you have at least drafted theoretically that, Hey, this is how my transfer from internal to external takes place. And this is how my new SOP may look like.

Then the next step is, which is step three, is to now start working on reducing your internal, because see, when is it that we are doing some value add, when the machine is running and for that, we need to start squeezing that pie of the I portion.And then there's the fourth step, which I've never gotten to. But theoretically, after when you have optimized your eye, you're supposed to go to E and optimize your E, those are the external steps. So these are generally the steps that you have to take for an effective changeover. And just to, give a more general example, coming back to the pit stop. If you observed the videos from 1950s or sixties, you'll see that whenever there was a pit stop, it used to take a lot of time. Don't quote me on it. It was more than 10 minutes or something, but recent videos if you see, it's less than five seconds, and you will see that there's no one team working on it. One team, what that means is the tire team or the fueling team. That's actually a system approach, means all the teams have to coordinate effectively for that changeover, for that pit stop to be more effective and to bring that time below five seconds.So the first step of a changeover management or SMED is to observe and analyze the process, to identify the internal and external process steps. Some people call it internal or external elements.

Siddhit: So Prashant that was fantastic. I really enjoyed how you explained it. I liked the way you went to the specific example in that manufacturing role that you had, but you also took it to a general example that was easier to understand which was the recent pit stop. So let me just kind of summarize what you said and I've spoken about this on the podcast in earlier episodes about SMED which is Single Minute Exchange of Die. And in this case for SMED, you have used DMAIC as the underlying framework, but for your approach or your module, you had four clear steps, right? And the first step was observe and analyze the process, to know what the internals are and the external side, and the way you define the internals and externals were, that the internals were things that could happen only at the time of the changeover.They could not happen at any time before they were, you know, serial, or they were very, very dependent. And the external is were things that could be taken out of this changeover and could have been done beforehand, could have been done in parallel and so on and so forth.So the first step would be to identify and observe the process to identify these two. The second would be that we would try and transfer as many things that we thought were internal and think about if they can be external. So to take an example from, like a changeover of dies, a very general example would be that if I thought that a new dye had to be made ready and you know, cleaned up or tuned for, or calibrated for the process, I don't have to do it while the changeover is happening.I could have done it like a long time back, or I could have kept it ready. So that would have then become external, right?

Prashant: Absolutely.

Siddhit: And then the third step is where it starts getting difficult, which is now we have to reduce your internal. So you've converted everything that you taught were internals, but are actually externals. And you're happy about that. So those can be done in parallel, but now really reducing the internal is where a lot of the coordination happens because it affects other things. It affects your standard operating procedure. And this is a point where this whole thing becomes the low hanging fruits are probably gone. And now you're really reaching for a lot of effort for some improvement. So you are now trying to reduce anything that has to be done during changeover, by reducing your internals. And fourth is like you said, you didn't probably reach to this step. And this could be, you know, also very difficult, which is optimize your external, which is what are the things that I have already transferred to my external, always external, but I can still optimize a lot of them. So those are the four steps. So Prashant, do you have anything to add to that? Great example, really enjoyed it.

Prashant: I think Sid you explained it beautifully. I can just, you know, add for the clarity of our listeners another example, which is much more general if they can associate it, associate better with this one. Say, for example, our airplanes. We always see a changeover happening when we are boarding a plane. Here in this case, the plane lands, comes and now the passengers have to get out. At that time our plane is actually on a standstill, means it's not actually transporting anybody yet. So the passengers are coming down. This is what you're talking about, the internal, anything that will go on on that plane at this time will be considered internal. The passengers are moving out. The cleaning crew goes in and starts cleaning. The fuel up crew goes to kind of fuel up the plane.The baggage kind of taking care of, also the maintenance crew is sort of going their own ways and so is the food and beverage and then there's the boarding process. So all of this is in a way, in that I category.

Now, the external would be, if I am taking, say too long to package my food module, which has to be uploaded onto the plane, but it's already packed and ready when the plane is here. But when I was actually packaging, it's taken me longer than normal. So that is optimization. First, we think about optimization of I, as you mentioned that, Hey, let's first reduce this changeover so that the passengers don't have to wait, the plane doesn't have to sit here for a long time. And then we'll think about the other stuff, which is later on, as you said, you know let's first take care of the essentials. And once we are done with that, we'll go to optimizing the the other layers of the onion. And I, as you mentioned, did not have the opportunity to kind of, you know, get that far yet.

Siddhit: Yeah, that was another good example. I've never thought of airplane changeovers, even though they happen in front of me all the time as being in the same framework, but I see it and I hope the audiences saw it as well. So thanks a lot Prashant. Let's get down to, our next question, which is, if you had a magic wand to change one thing about like the way you work or your role, or the manufacturing industry in general what would that be?

Prashant: That's a really good question Sid. The way our human nature works is, when we are working for a specific department, slowly and steadily we start to kind of have those blinders on where we are thinking essentially within the boundaries of the department itself. But when we think about it, the organization is not a stand alone department conglomerate. It's more of how the systems are working. It's how that department actually fits into the bigger picture and how synergetic things can happen between the departments. So if I had a magic wand and as I mentioned earlier, that I have a great passion for cross-boundary collaboration, my magic wand would want to kind of reduce these walls of cross departmental silos. I would love to see people not thinking in terms of the departments, but thinking in terms of the organizations and its common goal. And that way I feel that we can be more effective and efficient.

Siddhit: Yeah, that's a great answer Prashant. A lot of the speed is lost because of the loss of communication and the loss of translation, the translation losses between different departments getting siloed, and when all of them need to have their stake well defined and that involvement from the start, this often doesn't happen. And I think that's a great answer for the magic wand question. So thank you for that. And yeah, go ahead.

Prashant: So I would just like to add, with the walls coming down, one of the major challenges that most of the organizations also face is off the alignment, means the top management is generally sure of where we are heading, but are the people sure of it or not or are they actually aware of it or not? So with this cross-boundary collaboration, et cetera, the alignment also becomes much more effective or easier given there are always those challenges at hand.

Siddhit: Yeah, definitely.So finally, concluding like the last question, which is like a fun question, if this was 2051 and you were able to go into a factory, or if in 2051, your grandchild were to be in a factory, what would they see? What would the factory of 2051 look like?

Prashant: Yeah. So you're asking a very innovations type of a question or a creative type of a question from an engineer. So I'll try my best. What comes to my mind is that if I would go into a factory or if my child would go into a factory in 2051, they would see first a lot of automation. Secondly, what they would see is reduced inventory sizes, because I would think about you know, reduction in lead times, I would think about faster changeover times. I would also think about seamless interfaces between all the systems that we have in place. Currently one of the challenges that I see within the organizations is that we have different solutions for different aspects. For example, there'll be a different software leveraged for production and then a different one for inventory control. The interface where the things sort of meet, that's where it becomes very much of a challenge. And I would see this getting resolved a little bit on a more effective scale going forward. So yeah, in a nutshell, I would see, I think shorter lead times, a lot of automation, effective interfaces.

Siddhit: Yep. That's a great answer Prashant, and this is not usually often what people say. They're usually talking about like advanced manufacturing kind of concepts, but this is a very important point that you brought up, which is, you know, you can have very advanced things like, you know, 3d printers or ARVR, but if you're still going to have the same losses in revenue and the same bleeding that happens due to changeover, that happens due to any of the modules or waste, then that doesn't really help. Those things have to be also reduced to as much as they humanly can and with automation as much as it's even physically possible to reduce them. So I think that's a great thing that you've pointed out, which not often many people think about for the factory of the future. They're always thinking about the fancy or shiny objects, but not about these tiny, tiny things that bleed revenue. So great answer Prashant, and thank you so much for being on this podcast and, and talking to me. And, you know, even though you're so busy with the newborn and doing a lot of other work that's going on. So thank you so much for giving your time and giving us all of these answers and sharing that experience today.

Prashant: Pleasure is all mines Sid. Thank you so much.

Siddhit: All right. So take care. And I hope to have you again on this podcast in the future and stay safe.

Outro:[Background music] If you enjoyed this conversation, please subscribe to the means of production podcast, for more stories from people behind all the manufactured goods we use love and depend on. This episode was made possible by Pashi, the operating system for manufacturing. Pashi unifies the entire production process for any product and combating operator instruction and data interpret interfaces, stage logic and parameter thresholding, machine interfacing and configuration, robot board programming in coordination and stage to stage to production flow control into a single Pashi program. Check us out at Pashi.com. And until we meet again, have a fantastic day and take care.

Music from Uppbeat: "Falling" by Zayner. License code: SYFMAQN9WWBAPCSU


Help us get to know you more