Episode 40

Pharma Operations

Category :  Manufacturing & Supply Chain

In this episode of the Counterpoint Podcast, we bring into focus the challenges that the pandemic has caused on the already disrupted supply chains of pharma companies in India and across the world.

Achal Pande first gives us an analogy from the way train networks function to explain the real challenge with pharma operations. This makes it super-easy to understand the core problem in this environment. He then discusses how all the challenges can be sustainably eliminated to improve lead time and increase output. This will enable companies to respond with agility to changes in demand. Tune in!

Read more https://www.vectorconsulting.in/big-idea/accelerated-pharma-manufacturing/

Transcript
Shubham Agarwal : Hello and welcome to the Counterpoint Podcast. I’m Shubham Agarwal and today we are going to be discussing about pharma operations. The outbreak and spread of the novel Coronavirus has led to a dramatic rise in global demand from pharmaceuticals. As India is one of the top global drug suppliers of generics, it is being approached with sizeable orders by many countries including the US, Australia, Brazil and several South Asian nations as well. Now, in order to meet the rising wave of both the domestic and the export demand, Indian pharma industry will have to ramp up capacity. But I think this is easier said than done. In addition to the epidemic, there are many causes of small and large urgencies from the market, the discounts of stoppage of supply by a competitor, seasonality, and so on. But it is not simple for pharma companies to respond to these, since there are many challenges that are endemic to this industry. Because of which supply chains of duck products can takes six months or longer. Now, the pharma operations have their own nuances and are quite different from conventional manufacturing operations. These operations are often characterised by very high lead time coupled with very high variability. To offset high and variable lead times pharma companies maintain huge inventories in their warehouses to protect custom delivery dates. this leads to high inventories and higher cost of operations resulting from capacity wastages. So in this podcast, we shall try to explore the world of pharma operations, its challenges and obviously the ways to overcome the same I have with me, Achal Sharan Pandey, partner at Vector Consulting Group who has worked extensively in the pharma operations and has design solutions to overcome strategic and operational challenges plaguing the industry. So, Achal welcome to the counterpoint podcast. How are you?
Achal Saran Pande : Hi Shubham. I’m doing fine. And it’s great to be back again with another podcast
Shubham Agarwal : Wonderful wonderful. Achal the previous podcast with you have been in the domain of automotive & engineering. So, why don’t you help us in telling us how the pharma operations are different?
Achal Saran Pande : Okay, see, pharma operations are quite interesting as they are very different from the conventional manufacturing environment. These operations are characterised by dedicated routings. So, if you consider a typical value stream, you would observe that every product or every formulation goes through a fixed route, the route is defined by a fixed set of machines through which the product flows or the you can say product moves across various processes inside the plant. Now compare this with manufacturing operations in other industries. In other industries there is a generic routing inbuilt where there is a flexibility to choose specific machine.
Shubham Agarwal : Okay, why don’t you give us an example? Yeah.
Achal Saran Pande : Yeah. So, for instance, if I consider or if you consider a broad process flow diagram of a typical pharma operation, it will have the following five six processes, the first is dispensing of the product, the second is granulation, third is mixing, fourth is compression, 5th is quoting and 6th is packing. This is a very broad process flow. Now, what happens in this process flow is a particular route of a product would incorporate a fixed set of machines for the formulation. Say, for example, if there is a formulation f1 If we consider post dispensing, it would flow through equipments say G1 in granulation, it would then flow through equipment M2 in mixing, C5 in compression, maybe CO6 in coating and P1 in packing. Now, this G1, M2, C5, CO6 and P1 are machine numbers. So, this particular product would flow through a fixed set of machines. So, if you want that this product should flow through a granulation machine G2 in granulation, it is not possible, it has to flow through G1, right and this is a predefined route, which will not change. So, in a pharma plant, there are you can say there are as many routes as the number of products and therefore the complexity. So you can say a formulation like fenofibrate would have a very different route as compared to say Tadalafil or a Montelukast.
Shubham Agarwal : Okay. So, these are different basically products you’re naming.
Achal Saran Pande : Yes, these are some of the products, common products that are made by pharma companies in India. So, the idea behind is that the routes are fixed because of predefined batch sizes, then there are export related quality commitments and regulatory controls of drug authority and that is why it’s a fixed route. If if this is the case, then you can say that such kind of operations have very little flexibility. The flexibility is very limited. And if I have to put in simple words, these operations can be christened as rigid operations.
Shubham Agarwal : Okay. So Achal, if I if I tried to understand this sounds a little complex to me. Now, if there are as many routes as the number of products or like you said formulations, then there is definitely a possibility of, you know, many routes converging onto the same machines or equipments, isn’t it?
Achal Saran Pande : Yes, that’s correct. I will take a very simple analogy to understand
Shubham Agarwal : Yeah, I think that will help Yeah.
Achal Saran Pande : So, you can visualise a railway station or a normal Railway Station since you have you you I think you’re from Delhi, right.
Shubham Agarwal : Yes.
Achal Saran Pande : Yeah. So, you must have seen the Delhi railway station it’s so big right. And there are so many platforms in the station. Now, imagine that from each platform a route starts right. And from if the routes start from each platform, you would see that after a certain distance all these routes converge into maybe one or two tracks.
Shubham Agarwal : Correct, Correct, Correct.
Achal Saran Pande : Which ultimately leads to your destination. Now, you assume that there are multiple trains on these platforms and you release all of them at the same time right. So, what would happen
Shubham Agarwal : Yeah there is bound to like accidents or you know, collisions happening,
Achal Saran Pande : Yeah, the collisions happening, the best case would be that all of them are released from the platforms and then most of them wait over there just waiting for one of them to take the destination route, right, that is the best case otherwise every train would actually meet with an accident right.
Shubham Agarwal : Yeah.
Achal Saran Pande : So, so, pharma operations are very similar, you can say it’s like a network of tracks, which converges to certain points. And if you release all of the material through different tracks, at the same time, they will come at the conversion point and wait for quite a long time
Shubham Agarwal : Okay.
Achal Saran Pande : So, if I have to say take a typical example, in pharma operations, you can say routings of formulations F1, F3 and F6 may converge at compression machine C4. Similarly, some other formulations, say maybe formulation F2 ,F10 or F15. I’m just taking some terms right may converge at some at say coating machine C2 and so there are many converging points. So, therefore, many routings converge at many multiple points in the plant. The problem with convergence is that there is a possibility of machine getting overloaded from time to time. This happens when the products are released or dispensed almost at the same time without considering the load on the converging machines right. So, if you remember this analogy of the track analogy, right, if I don’t look at what is the load on those one or two destination tracks, right, then obviously, you will meet up with an accident right. So, similar is this case. So, when these release products reach these converging points at almost the same time, waiting time gets built up as the machine is already pre occupied by the previous order or previous formulation. Now, if there are multiple converging points in the process flow, you can assume a significant waiting time getting built up, which is a summation of incremental waiting times at all converging locations.
Shubham Agarwal : Oh right, right. I think the the analogy is really helpful, because suddenly I understand the entire thing. So as I rightly understand, convergence may lead to route overloading, which in turn induces high lead times right.
Achal Saran Pande : I would say I would go a step further in this, see converging points or convergence is given to us it’s inherent in the process, okay, the problem is the way planning is done, how planning manages these converging points that is the issue. So, the way planning is done creates waiting periods at these converging points which leads to route overload. So, I would again consider the same railway track management analogy right. Imagine if you have a good signalling system which we have you must have noticed also right yeah, which ascertains the load on the tracks and schedules releases accordingly this problem will get solved and in fact, this is exactly how our railways track management system works. However, in pharma operations, the daily releases on the shop floor are based on monthly plan and not based on daily load assessment of routes and therefore, they grapple with high lead times leading to inventories variability and poor on time performance due to overloading, this is the you can say the issue
Shubham Agarwal : Right and I think clearly the pharma industry can learn a bit or two you know from the railways, which is So, weird, but that you know you can get lessons from anywhere
Achal Saran Pande : In fact a very good analogy I mean a way of looking at things differently.
Shubham Agarwal : Yeah, suddenly the entire pharma industry is quite clear to me how it works from the inside
Achal Saran Pande : Yeah.
Shubham Agarwal : So, Achal based on the railway analogy itself, you know, you had mentioned that do you want to say the pharma operations also then require a scheduling and signalling mechanism instead of a monthly scheduling mechanism like they have right now.
Achal Saran Pande : Yeah, precisely, we require a very robust, dynamic, scheduling and signalling system, a mechanism which controls releases on different routes based on load assessment of each route that is what we require . So the real change in paradigm is to move from fixed schedule dispensing to daily dynamic dispensing based on signaling of queues
Shubham Agarwal : Right right. So, like you pointed out, you know, that the releases as per the monthly plan may create the overloading of certain routes, does it also result in under loading if I go the other way around of the same routes
Achal Saran Pande : Yes, it happens the other way also, I mean, you may find it very strange, but it happens see, what happens is when the releases are are as per monthly plan, overloading and under loading of routes happen at the same time. See, overloading as we have discussed would lead to queuing of orders and in front of certain machines and would limit the output of the plant because the orders would go in but they would not come out at the same rate, right. So, this would increase your lead time at these work centres. In the absence of load monitoring some routes also remain in under loaded condition because what happens is I am releasing orders as per monthly plan right. And if I’m reading orders as per monthly plans, some routes may not get any orders at all, this leads to under loading of certain routes and under loading of certain routes would lead to starvation of machines or capacity wastage. So, what you see you have two phenomena happening at the same time, one route is getting overloaded and one route completely starving or you can say some machines are overloaded and some machines are starving. And therefore, what you would observe it’s a predicted effect that you would see in the pharma industry that even for incremental output pharma plants plan capex without realising that there’s enough capacity already available with them
Shubham Agarwal : Okay, oh.
Achal Saran Pande : So, I will again take a very simple example you would remember there used to be a saree ad coming in sometime back “Saree mein saree”, right. And and a car ad also said more car per car, right. So, I would say in pharma industry, there is a plant in a plant, there is so much of hidden capacity, but all the capacity is getting wasted. So, if you look at pharma plants very closely, you would find a plant already hidden within the plant, it’s a plant in a plant scenario.
Shubham Agarwal : Oh okay, strange and very interesting, I think.
Achal Saran Pande : Yeah,
Shubham Agarwal : Right. So, you know, the overloading and the under loading of roots would then limit the output of the plant, you know, and generate high lead times obviously. So then how do the planners manage this capacity to ensure that, you know, the due dates are met and wastages is reduced? Because I’m sure this has been going on for quite some time.
Achal Saran Pande : Absolutely. See, there are two strategies or tactics, the planner or the production people adopt, the first tactic they adopt is to protect the lead time. Now, you see as the lead times expand, right, because of waiting at various work centres, the delivery schedules or the delivery dates are not met as per customer expectations. And therefore, the need to release the order much in advance increases, right? Yeah. Right. So you would observe that the planners, what they do is, they start operating with a higher order visibility of three to four months. And even in some cases, it’s six months. Right, so that they can release much in advance and meet the order due dates. They actually want to offset higher lead times by releasing the order in advance. But remember what I said before, the more you release, without assessing the load across equipments, or on routes, the more the lead time.
Shubham Agarwal : Yeah, it is only going to increase correct.
Achal Saran Pande : And therefore it’s a vicious loop, which gets activated. Where an increase visibility generates cueing. The cueing generates higher lead times, which further reduces the need for more visibility, it’s a trap. And many fall for it. So this vicious cycle, when it gets activated, it only generates more and more wastages.
Shubham Agarwal : Okay.
Achal Saran Pande : And you get nothing in return.
Shubham Agarwal : And what’s the second tactic?
Achal Saran Pande : The second tactic deployed is for protecting the capacity. Now, the setup times in pharma across machines varies anywhere between 16 to 24 hours per setup, which is huge. So, what they do is in order to protect this capacity, the planet generates campaign of products or formulations. Now, we need to understand what a campaign is
Shubham Agarwal : Yeah, I was about to ask what a campaign is,
Achal Saran Pande : Yeah. So, a campaign is basically clubbing of multiple batches of the same product. So, if you if you say that a campaign of three then a campaign of three implies three batches clubbed together and released in the system, okay. Similarly, if we say a campaign of six, it would imply six batches clubbed together and release in the system. Now, when the planner has long visibility, he gets this flexibility to pull ahead future orders of the same product, prepare a campaign and release them onto the shop floor. But you see, when a very large order enters the system, I would say it’s like an elephant order entering into the system, elephant order is essentially a very large order which eats up a lot of capacity in one shot, so, it not only creates waiting time for other orders, but also creates capacity wastage. As some of the club orders are not immediately dispatchable to the customer. This we call as capacity stealing in pharma operations, because you are stealing capacity with such a large load right. So, you see another vicious loop comes into play, a second vicious loop and this vicious loop is related to capacity. So, what is the vicious loop? The vicious loop is protect capacity by releasing larger campaigns to reduce setups which leads to capacity stealing, which is a capacity wastage. So you generate more campaigns
Shubham Agarwal : Okay, okay. Right. I think the the problems that you know mar the entire pharma operations are very clear. And you also talked about the signalling mechanism for releases to prevent queuing before work centres. Are there additional solution elements that we may also look at to resolve such a situation?
Achal ?
Achal Saran Pande : Yeah, in a in a nutshell. There are three tactics which have to be deployed. The first is that we have to look at establishing a daily scheduling and signalling system for controlling order releases on the overloaded routes and releasing order on under loaded routes.Once the releases on the shop floor are dynamic the sourcing of raw materials cannot be schedule based. It also has to change to a dynamic sourcing mechanism We also have to look at as a second tactic to reduce waste, such as setup times or major breakdowns. If there are any, what also happens is in certain environments, it has been observed that the cycle times are also inflated, cycle times of these machines, which may be relooked or evaluated to bring down the process times. So objective evaluation of machine speed keeping quality norms intact, has to be evaluated to achieve higher production and desired quality levels.
Shubham Agarwal : Okay, right, right.
Achal Saran Pande : There is one more area that we have to look at. But I do not know we can do it right now, because it’s quite detailed and complex, is we also have to look at quality testing times in the QC labs. And why I am saying this is because after every process, the product goes for QC testing.
Shubham Agarwal : Right, right. And there’s an elaborate process.
Achal Saran Pande : Yes. It’s a very elaborate process. The turnaround times are high. And there are various a lot of wastages are like OOS out of specification, OOT. Right. And incidences, right. So these wastages have to be arrested to bring down the time of delivery to the customer and also release capacity.
Shubham Agarwal : Okay, okay.
Achal Saran Pande : So this is a something we may discuss in some other podcast, because this is an area by itself, which needs to be addressed the QC lab part.
Shubham Agarwal : Right, definitely Achal, I think we would love to discuss this in detail, because I think the QC part is definitely important for any pharma operation. And since you said that, you know, it’s it’s an elaborate process. Let’s come back and discuss in another podcast, how do we tackle this part of the, you know, the operations and how do we release capacity from there? Great, I think thanks a lot for the discussion Achal. I feel it was really helpful. And I think the analogies and examples that you gave us, made it very easy to understand as well. And I’m sure the listeners would appreciate that. For all the listeners. If you have any questions or concerns on this topic, or if you want to have any more thoughts on this, please write to us on our social media handles. The link is in the details. Thank you Achal. Once again, we’ll come back in another episode and discuss on QC.
Achal Saran Pande : Thanks a lot, Shubham and I look forward to it.
Shubham Agarwal : Thank you. This is Shubham signing off. Until the next episode, bye.
Comments

Your Comment

Your email address will not be published.