welcome to another edition of a firmsconsulting case video today we're going to look at operations cases and I must say that invariably I find most candidates struggle with operations cases it seems that everyone wants to be a strategy consultant everyone spends their time looking at strategy cases such as how do you enter a market what do you do if a product line is failing how do you deal with the deregulation of the market and while those are exciting cases and to be fair I used to be a corporate strategy consultant so you know I do find them interesting the reality is that operations cases are an essential part of case interviewing and you have to understand them and to be honest operations cases are more difficult than strategy cases because in strategy cases most people can figure their way through them but in operations cases unless you really understand the language and you have some context you are going to fail so what we're going to do is we're going to present an Operations case here that tests whether you understand the basics or the first principles of operations if you don't understand the first principles of operations you would not be able to solve this case and invariably you would not be able to solve the even more complex operations cases and as I mentioned many times and as this case is going to show operations cases are numerically challenging because you tend to be dealing with unit allocations and little tiny numbers that you need to add up across now the thing about operations cases if you do want to learn about operations cases you can go read several books a book that we recommend is the book goal by I think it's dr. Eli Goldratt you don't need to read the whole book I think it's about 300 pages you could get a summary and there are many summaries available at that book in fact the two to three page summary just explain the core concepts would be more than enough for you but that is for a different day today we'll assume that you have an understanding of operations so you can maybe go ahead and pause this case and if you don't have an understanding we'll teach you the basics of operations the case we're going to do is a McKinsey case it's rated at the difficult level and it takes about 30 to 40 minutes to solve this case the math should come quick because it's basic math but the logic of how you're going to do these things should take a little bit longer the case is Pepsi has one major Balkan plot in Los Angeles in last three month months inventory as more than doubled while customer complaints that's the people that Pepsi Bottling Company is delivering the cans of Cola to or the bottles of Cola are complaining because basically we don't know why but their complaints have gone up fivefold so what so the question is what would you do well in a real case the first thing you got to do is understand all of the information you've been presented with so for me personally the thing that stands out there that I'm not very clear about is when they talk about customer complaints they haven't explained why there are customer complaints right and as you know as we teach our candice if you experience or you've given a lot of data the odds are pretty good that they're expecting a framework before any clarifying questions right that's - pretty much my McKinsey and Baines handle cases and while this is the McKinsey case I've kind of presented it like a BCG model with very little data the conceptual issues are hidden and you've got to dig it out in a McKinsey case they tell you really really clearly this is the only problem there are no curveballs move ahead with the case right so let's now explore this case given the fact that so little information has been presented to us I think it's fair to say that we can ask additional questions but even so just to be fair I would always check with the interviewer whether he wants my clarifying questions or where they wants to see my approach first let's assume this is a friendly guy he had breakfast this morning and he's really happy to have a good chat with you so what are the questions I would ask him well firstly inventory could mean many things is it finished inventory that's sitting in a warehouse or is it inventory in progress that's an inventory that's arriving from suppliers very important point because depending on the inventory you have it's either a delivery problem or it's a production problem because inventory in progress is coming into the manufacturing process finished goods inventory as left the manufacturing process or says why am i wire customers complaining that's an obvious one I think many of us would have picked that up if we were doing this case I'd like to know about how much we want to reduce inventory and customer delivery times I think that gives us a target we're trying to chase and as you know most candidates do struggle with cases because while they do the case as well they don't know where the deadline is if you know about how much the interviewer wants you to reduce inventory and customer delivery times you know when to stop the case when you reach that target then obviously I'd like to know are there any other objectives noise profitability return on investment return on invested capital are there any other targets we need to worry about and of course if there's more targets than just inventory in customer complaints the case becomes much more difficult because the more targets you are trying to chase the more things you have to keep in the back of your head so let's assume that we're told that all of the inventory some supplies that's coming in before the manufacturing process customers are complaining because they're you know delivery times have gone up dramatically triple to gone up to about five times inventory reduction and customer delivery time reductions as much as possible and there are no other objectives what do we do now well the next step should be quite obvious now that we've been given all this information I would take a few seconds I tell the interviewer I take a few seconds to gather my thoughts and I would present my key question how can we reduce inventory costs and customer delivery times at the LA bottling plant now the kind of framework we give you here is a framework that if you presented it to an interviewer he would be unhappy so listen very carefully what I'm going to tell you next he would be unhappy so that you need to handle this in a certain way so let's present the framework and show you how you handle it first you need to look at inbound logistics that is stuff that is coming into the manufacturing process we know there's a problem with supplies but we don't know if it's a problem with suppliers right there could be a parameter supplies on our side in the way we order or they could be a problem with supplies on the way they fulfill our orders manufacturing we've been given no indication that manufacturing is a problem right so for now we just assume there's no problem there and we know the in terms of outbound logistics we have some information saying that deliveries our problem for those of you are not familiar with this this entire three-step process is known as the supply chain if you go into an interview and tell an interview I'm going to analyze the entire supply chain that's my framework they will laugh at you or they'll throw something at you hopefully something that heavy so what you've got to tell an interviewer that preempt them say look I understand this is the entire supply chain by no means intend to say that this is my entire framework this is just some considerations that what I'm going to do now is identify which part I'm going to analyze and why that way we can narrow down our framework to just one point that we want to analyze and focus our case make sense let's continue the step from here is obviously asking a few questions to determine whether the problem lies in the inbound side on the manufacturing side or the outbound side and you've got to be efficient you can't ask a lot of questions you've got to ask one or two questions that eloquently and efficiently show you understand the problem so what would those questions be well firstly I would like to know why we are buying more materials well clearly there's some hold up there so why are we buying more materials has anything changed in the manufacturing process you'd be shocked how many people forget to answer ask this really basic question it just baffles me you know we've been doing this can't your cases with some Korean students recently and one of the things you've got to understand about the Korean education system is it's quite advanced but operations cases don't seem to be on the table on the menu for South Korean students they really good at strategy cases but they really mess up operations cases and I just did a session with about maybe five Korean students and all of them fail to get this issue now it's not just Korea that struggles yet because obviously an outstanding education system but we find invariably Candice from across the world struggle because no one is primed to think in operations terms outbound logistics I would like to know why customer orders are delivered late so look at the sequence here right we were told they are complaining then we found out why they're complaining because they are delivered late and now we want to know why they're delivered late so let's see what the interviewer tells us well he's going to tell me demand has gone up from six thousand bottles a day to ten thousand bottles a day since it is now summer that makes sense you know as demand goes up you'll buy more raw materials more inventory to fulfill demand they've changed nothing in the manufacturing so there's interesting because the manufacturing process for six thousand bottles is the same as the manufacturing process for ten thousand bottles not necessarily a good thing but a clue as well and customer orders are late because the factory is not producing sufficient Cola to meet demand and we have forced a can for shipment right that years some interesting clues for you that you can take and work with right and very basic stuff here basically due to demand an increase in demand more supplies are being bought but due to something in manufacturing those supplies are not being converted to finished products which is why there's a delayed customer so clearly there's a manufacturing problem we've got enough inbound supplies coming in but they're not being converted in the manufacturing process and obviously the outbound process is failing because there's nothing to take from manufacturing to customers so let's explain this with some numbers right let's look at the former state before demand went up demand was six thousand bottles a day right so my hypothesis is that due to a bottleneck or other manufacturing problem this is causing an inventory buildup and manufacturing shortfall which is leading to customer delays and notice the fact that I've got three parts to my hypothesis and I pollicis can never be that there's a manufacturing problem that is an hypothesis that belongs in a non accredited college if you go to an accredited college you should understand lipolysis must have three parts you must have the observable phenomenon the cause and the effect if you gave me an iPod thesis with just two of the three parts I'm okay with it but if you give me a hypothesis that cannot be tested if you just tell me something like there's a manufacturing problem and you cannot test it that's never going to fly so let's look at this in a graphical layout so let's lay out the manufacturing process let's assume there's just three parts to the manufacturing process right the first part is where you get the bottles and you sterilize it and let's assume the capacity of that plant for a day of that part of the plant per day is 10,000 bottles a day filling of the bottles with Cola the capacity is 9000 bottles a day and the bottling when I mean bottling I mean putting a cap on those bottles is 8,000 bottles a day right so let's look what happens on a typical week you've got six thousand bottles six thousand bottles bought because that's the demand per day the product deficit that means that the product the demand that is not met is zero because all six thousand can go through the production process because the bottleneck is above six thousand and that happens for the whole week from day 1 to day 2 to day 3 today for today 5 so everyone's happy all the customers are happy right but what happens in demand close to 10,000 bottles a day well the bottling side where you're putting a cap becomes a bottleneck because we've demand exceeds supply you have a bottleneck and if you have a bottleneck it's automatically a production problem so now we know we have a bottleneck and what does that mean in terms of happy or happy customers well lets them erase this and see what that means so we now have 10,000 bottles of inventory being bought right that's logical right what is the product deficit well in the first day you can have 10,000 bottles going into the manufacturing process and being queued up only 8,000 bottles are going to come up so in terms of the product deficit demand not being met we have 2,000 bottles not being met right make sense let's go to day two right in day two what's going to happen well in date two what you have is that from day one you've got 2,000 bottles that have queued up I was in excess inventory so waiting to come in and then in day - you've got the $2,000 bottle deficit from day one plus the two thousand two thousand bottle deficit from day two giving you four thousand bottles of demand that are not being met day three same thing right in day two you now have four thousand bottles queuing up plus the two thousand that are going to come in from day three you've got a six thousand bottle deficit and you can simply repeat the process now what happens when you repeat the process is you find out that this very innocuous bottleneck problem is causing a dramatic queue whereby by the end of the first week you're going to have 10,000 bottles queuing up and if 10,000 bottles are queuing up for manufacturing that means 10,000 bottles that need to go through manufacturing and not going through manufacturing and you've got a product deficit of 10,000 now that we've outlined this problem hopefully a little bit eloquently what do we do with this well if you've identified the problem the next step is obviously to generate options to fix said problem you have two options here the first option is you can fix the bottleneck so that all the supplies are used and there is no product deficit and/or inventory buildup now McKinsey particularly and I would say all firms want to see creativity I get really bored of the case when someone is just throwing out the most basic stuff right so let's look for a really creative answer yeah we know this is a seasonal problem because during summer the problem occurs so another more creative solution is that we shift production to a neighboring water like maybe I don't know in the VAD I think it's close to California and we import the product deficit into Los Angeles we clay could even shift production to maybe a neighboring bottle maybe in San Diego San Francisco somewhere within California but the point is you don't have to necessarily do any changes to the LA plant now obviously here you gotta prioritize which option do you analyze first right once you've presented your options you've got to analyze them you've got to analyze them on certain criteria and decide which option you want to pursue dealing with your own plant is probably the best thing to do because potentially there are ways to fix your bottleneck without incurring significant cost you want to look at that first and only if that's not a solution then you look at making massive capital changes or shifting production so let's explore that let's take this even further right so fixing if you want to fix the bottleneck we have three options we can move quality lashawn's before the bottleneck I can show you right now when I tell people this is a potential solution they look at me with a blank face or they look at they look at me like I just told them the earth is actually flat and it's not round expanding the bottleneck is an obvious solution that's what most people say and sharing the bottleneck work with another part of the plant they may be another part of the plant who could do the same work as the bottleneck with a few changes so this is creative and it's usually possible because what happens is plants do have spare equipment or they do have other equipment that can do the work of the bottleneck so now let's look at the one about moving QA before the bottle like I'm going to explore this in more detail because it's the one most people to understand the least but of course all of them you have to explore right you have to look at the criterion what is the cost-benefit analysis of doing each to come up with a recommendation but we just going to look at the Q&A one because I think it makes the most sense here so let's look at moving Q&A before the bottleneck now what does this mean right let's go back to our plant we've got three steps in the plant sterilizing the bottles filling them up with Cola and I'm putting a little cap on them on the day one we've got 10,000 bottles if the bottleneck is 8,000 we assume 8,000 bottles will come out but it's not true the apparent capacity of the plant is only a hundred percent that is to the apparent capacity is 100 percent in reality you are going to have some defects occurring let's assume the defects are 5 percent right which is 400 bottles the real capacity of the plant is 95% and if you spread this across you know you're losing $400 a day for the plant right so 5 percent of products going through the bottleneck will be rejected meaning the bottleneck is not working on hundred percent capacity but rather just 95 percent capacity so what you can do is you can see here that if you look at the losses they're quite significant they keep on adding up every single day right it's four hundred a day if you can move the queue a step before the bottleneck right you would remove these products before they entered the bottleneck so rather than sending only seven thousand five seven thousand six hundred good products to the bottleneck you would send eight thousand products to the bottleneck because the products that we're going to be kicked out after the bottleneck would be kicked out before the bottleneck thereby only sending good products to the bottleneck what this means is that if you move the QA or part of the QA before the bottleneck you ensure maximum product is produced but now the case you have to ask yourself is well you can fix this and it's quite easy to fix should you make capital expansions to address the 1600 bottles that you still need to find remember demand is ten thousand if you've found another four hundred you've got eight thousand four hundred moving through which means you've got one thousand six hundred bottles that are not moving through now since this is seasonal importing seems to make more sense to me I would actually say why go through all these capital changes when just for four or five months of the year we can get another plant to work it may be slightly high capacity without any chain just to their production right so let's just recap where we've come with this case well you've got to tie up the case right and that's what you and I mean what's what I mean when I say recap this is what I'll do I'll tell them interview that in this case the inventory build-up and delays and customer deliveries are driven by the same problem right it's basically a bottleneck in the bottling step of the bottling plant and they fix the problem we can do many things but the simplest thing we can do immediately would least amount of effort is move QA before the bottleneck we can improve throughput capacity by five percent and we assume it'll cost us nothing right this means that we still need to find another one thousand six hundred bottles per day right and there are two ways we can address this we can introduce capital expansion into the plant or we can shift capacity to another plant that may have idle capacity right and of course comparing the capital side of things capital changes and conveying the option the broad option of shifting capacity to another plant will have to do a cost-benefit analysis but all other things being equal shifting it because it's seasonal and it's going to come back to normal shifting it another plant would tend to make the most amount of sense to me so now that we've covered this case let's sort of recap some of the key things you need to know the observations here is that in the supply chain everything needs to be broken down into three convenient steps but you've got to understand that this is a very broad process and if you just tell the interviewer you're going to go with the entire supply chain they're not going to be happy so you have to quickly follow that by saying that I'm going to use a few sets of clarifying questions to pinpoint where I think the problem is and develop a hypothesis and that is where we will focus the analysis when you think of a supply chain you should think of a set of dominoes you move one part of the Domino there's a ripple effect across the supply chain your job is to find the relationship between these seemingly unrelated problems because here we saw that infantry was a problem and customers at no time did we think that manufacturing could be a problem until we understood the relationship of the different parts of production in a supply chain I'll tell you again operations cases of a numerical require lots of calculations and I love doing them with candidates in the final stages of the training one of the most famous cases we do is a Russian mobile telecommunications case looking at how dropped calls increases during the Russian which in Moscow peak rush hour between 6 o'clock and 9 o'clock failure rate on that case sets are closed about a hundred percent and I always find that candidates make mistakes there because of unit conversions you're dealing with bottles per day bottles per cycle set up a time number of people per hour labor cost per a unit of product you don't use units you are going to die for those of you who are in our training and have gone through our estimation cases you know how important unit conversions are in operations cases I think of a very big estimation case with a lot of different equations equations without units is language without structure no structure no understanding everything fails operations cases can always be broken down to three steps if you don't understand bottlenecks and operations cases you are dead understand that process the principles of solving operations cases for those of us with those if you've been through the training you know we we take this directly into estimation case when you look at supply and demand we're taking the telecoms cases we take it heavily into IT cases the point is knowing operations cases is not just important for operations cases it's important for all kinds of cases now if you really want to improve your knowledge you should understand marginal cost analysis because I find that when we start when we when we give you a case with an Operations concept and marginal costing I would say most candidates fail them most candidates really struggle to understand this concept and you should look it up because in future cases that you do with us we are going to throw this concept at you these cases tend to be very long there's no way you're going to be able to go through inbound logistics manufacturing outbound logistics on top to bottom you'd be there forever your job is to quickly pinpoint the problem let the interviewer know that you know the framework is too big and that you understand you have to pinpoint the problem right so let's just look at how candidates have done generally in this case so what we've done here is we've taken a case summary looking at how candidates have done across the last two years with us on this kinds of cases in terms of case skill requirement you have to be very good at case structuring and very good at communication to solve operations case it's very good a case structuring because if you have the wrong structure you are basically going around in a maze or you're gonna have to if you use the wrong structure the structure may tend out to be so cumbersome to use you are never ever going to develop it right you're never going to solve the problem the second thing is because you're dealing with so many units of calculations and bottles per day cost per hour units per second unless you are really communicating what you want to do why you want to do it how you want to do it there's a great chance of miscommunication I know many candidates will resonate with the point I'm going to make if you ask me the wrong question and I answer the question correctly you are screwed and I do find that with many candidates what they want to ask me and what they ask me is the interviewer is very different from what they intend to ask and if you ask the wrong question I give you the right information for the wrong question you're going to end up being totally lost in a case especially when you're dealing with these very nuanced units of conversion but how do candidates do well I would say that the reason this case is so challenging is because it expects you to understand operations cases at its core the case is not challenging but the principles it applies means you must understand operations and if you don't well you know what you did how do candidates do I find MBA candidates do okay in this case they tend to do well on business judgment they tend to be do well in case structuring because most MBA schools have taught in operations costs and a lot of candidates can bring that to the table but I can tell you right now all MBA students struggle and I don't think anyone's ever finished this case so while their business judgment may be strong they're not able to parlay that into communicating with me effectively to extract good insights from me as the interviewer but also from extracting guidance from me and case structuring our find to be quite a weak for MBA candidates what I do find is that undergraduates tend to be a weak on communication and weaken business judgment and they weaken business judgment here because I do find that undergraduate programs in business and generally history art scientists don't teach operations so operations is something you've got to read and the only reason I post sorry undergraduate students do badly it's because they don't have this operations background but I do still feel that our undergraduate candidates do better on the arithmetic and calculations in our MBA candidates and this is looking across the last year and sort of averaging out performance for all the candidates who have been through the program why will you fail this case you have week operations knowledge you need to get the summary of operations from any MBA textbook and you'll be fine as always I'll be happy to respond to any questions and I do hope that you found this case to be very useful to your preparation and training thank you
