Supply Chain Resilience Fails When Decisions Move Too Slowly

What if your biggest supply chain risk isn’t disruption, but the time it takes to decide what to do next?

In this episode of the Resilient Supply Chain Podcast, I’m joined by Robbert de Looff, Industry Commercial Lead for Chemicals at OMP, to explore why supply chain resilience now depends on more than better forecasting. In a world of energy price spikes, shipping disruption, raw material constraints, sustainability pressures, and geopolitical shocks, visibility is useful, but only if it leads to better, faster decisions.

Robbert and I break down why traditional planning cycles can leave companies reacting weeks too late, and why decision-centric planning is becoming so important for supply chain leaders. You’ll hear how organisations can move from rigid S&OP rhythms to scenario-based planning, where teams know what data they need, who owns the decision, and when action is genuinely required.

You might be surprised to learn that “real-time planning” doesn’t mean constantly changing the plan. Sometimes the best real-time decision is not to act. We also explore where AI can help, from surfacing relevant risks to running what-if scenarios, and where humans still need to stay firmly in control: relationships, judgement, and trust.

Kismet: one of the sharpest examples is the Rhine running low. Not a cyberattack. Not a system failure. Just water levels quietly deciding whether chemical supply chains can keep moving. Resilience, it turns out, can still be humbled by a river.

🎙️ Listen now to hear how Robbert de Looff and OMP are rethinking data, visibility, risk, and decision-making for more resilient, sustainable supply chains.

 If disruption hit tomorrow, would you know where your supply chain was most exposed? In 15 minutes my free scorecard helps you assess 27 resilience statements, calculate your score, and turn the result into three priorities and a 30 day action plan. You can download the scorecard free at tomraftery.com/scorecard.

Transcript

Copy

Robbert de Looff: 0:00

If you don't make the decision yourself, the decision is made for you. And typically not in your benefit. If you need to make, make a decision, do I gonna go for another supplier? Yes or no? If you don't decide, someone else did, and you can't even make the decision anymore 'cause it's a no.

Tom Raftery: 0:14

In supply chain delay can look responsible, but sometimes waiting for the next planning cycle simply means the decision gets made without you. Good morning, good afternoon, or good evening, wherever you are in the world. Welcome to episode 125 of Resilient Supply Chain stories and strategies that Keep business moving. I'm your host, Tom Raftery. My guest today is Robbert de Looff, Industry Commercial Lead for chemicals at OMP, and we're talking about why traditional planning rhythms are struggling in a world of energy, price spikes, shipping disruption, raw material constraints, and sudden geopolitical shocks. Because the issue is not just volatility, it's decision latency. Robbert makes the case for moving from process centric planning to decision centric planning, where companies rehearse scenarios, know who needs to act and can move before a disruption, becomes a margin problem, a service problem, or a full blown operational mess. Interestingly, one of his examples is the Rhine running low, not a cyber attack, not an AI failure, just water levels quietly deciding whether chemical companies can move product efficiently. Supply chain seemingly can still be humbled by a river. Okay, let's get into it. Robbert. Welcome to the podcast. Would you like to introduce yourself?

Robbert de Looff: 1:42

Yes. Thank you Tom. Thank you for having me. Very interested in the upcoming podcast. So my name is Robbert de Looff with OMP now for over 10 years. And I'm the Industry Commercial Lead for chemicals. Means responsible for all new sales deals coming in, in across the whole globe for chemicals, but also of course, our strategic accounts in chemicals.

Tom Raftery: 2:01

Okay, great. And for people who are unfamiliar, Robbert, could you give us a quick 101 on OMP and then tell us as well, what problem are you trying to solve in supply chain planning?

Robbert de Looff: 2:14

Yes. So OMP is an advanced planning solution advanced planning solution covering indeed your supply chain challenges, covering them from a full end-to-end perspective, end-to-end, meaning that it covers procurement, production, distribution, and demand. So really your full value chain, but also end-to-end, meaning that it covers your full horizon. So going from your network Rhine long term, 10 years out, closer towards your tactical horizon. So let's say a year, two years out, all the way down towards your operational planning and in the end, even your scheduling very close to execution. That actually already points out a bit what OMP tries to solve in it. It tries to indeed, optimally make a plan, make sure that you can make strategic decisions that flow true towards execution in the end. And that's also where our solution is built on one single core, that if you make a change in your tactical decision to fulfil a certain high margin customer, that it in the end results in this being executed.

Tom Raftery: 3:13

Superb. And What does your role give you visibility into that most executives don't see day to day, would you think?

Robbert de Looff: 3:21

What I think here is interesting to mention is often people have roughly in their mind what they need. They just don't have the capacity, capability, time, whatever the reason is, is to execute on it. So what we very often see is, is companies coming our way, is that they're lacking this end-to-end visibility. And they do know that they're lacking the end-to-end visibility. But how to tackle that and how to translate it into getting the actual end-to-end visibility, that's something that's typically not something they know or used to doing. Whereas of course, we have the experience in doing this many times. So what you often see is indeed breaking those boundaries between, for example, your procurement guy, and your distribution guy, and your production guy, that they work as a team together looking at the same scoreboard and not one is optimising to get the cheapest possible, but maybe with a long lead time. Whereas then production will say, I can't produce optimally because procurement is not delivering what I want. So I typically see, which others do not see is the approach of moving from this siloed let's say more offline planning, which is happening towards indeed this integrated planning all in one

Tom Raftery: 4:30

platform.

Robbert de Looff: 4:30

Both that horizon part, but also that let's say end-to-end supply value chain part.

Tom Raftery: 4:36

And when we talk about planning, does good planning need to be fast, or can it be slow or, what's a moment when a good planning process suddenly becomes too slow?

Robbert de Looff: 4:52

I, I like that question. And it's, it, I'm gonna first start with a typical consulting answer. It depends, but I'll elaborate as well. I think you need a certain structure in the first place. You can't just go outta nowhere towards immediate decision making. Why? Because it's not just a solution that's gonna help you do the decision making. You need the organisation to be ready. You need the processes to be ready. And, and this all comes together with the solution. For this one, I think you need to step by step, move into the direction you want to be. What is that thoughts on the horizon where you wanna be in, let's say five years from now, or one year from now? Actually, different dots on the horizon, so to say. And yes, speed is extremely important. I mean, if you look at today's world, I mean, there's so many disruptions. The Middle East conflict, the Red Sea conflict before of course also the war Russia, Ukraine, which had a, massive impact on Europe. And there there's many more across the whole globe happening. So you need fast decision making. But if you don't have a solid, strong structure and a solid, strong process to make those decisions, then it could perfectly be that you're not making the right decisions after all. Yes speed is important. It's getting way more important these days, but you still need to have that solid foundation before you can start speeding up.

Tom Raftery: 6:11

Okay. And would you say the real problem today in supply chain is volatility or decision latency?

Robbert de Looff: 6:19

They're a bit connected, I would say.

Tom Raftery: 6:21

Okay.

Robbert de Looff: 6:23

The volatility, of course is asking companies to make faster decisions. And there was one account in which a person said, well, if you don't make the decision yourself, the decision is made for you. And typically not in your benefit. If you need to make, make a decision, do I gonna go for another supplier? Yes or no? If you don't decide, someone else did, and you can't even make the decision anymore 'cause it's a no. So volatility for sure is pushing for more decisions. So the fact that now indeed prices spike or go down the fact that there's new legislations or trades regulations coming up, you, you do need to make decisions fast to make sure that you get an optimal plan. As a result, this volatility, these changes in geopolitical situations, prices even demands force for fast decision making if you want to get towards a optimal solution in the end.

Tom Raftery: 7:13

And a lot of companies would say that resilience means better forecasting, but is the bigger advantage now, faster decision making when the forecast breaks.

Robbert de Looff: 7:25

Yes, I would say yes. And actually I'm gonna refer to a colleague of mine who once that resilience is the agility you plan for. And that quote stuck with me'cause in the end, it's true. Everyone wants to be more resilient and there's definitely ways of being more resilient in your operations as well. Imagine something breaks down, how can I quickly shift around? But in planning in essence, resilience means that you have a plan ready that can cope with downsides or upsides, which you have not, let's say, predicted, or that you do not see as the average expected result or forecast that's gonna come. So if you wanna be resilient, you do not need to work with a single forecast. You want to work with a potential upside, a potential downside, a potential situation in which a Strait Hormuz stays closed, a little longer, a little shorter, etc. So you need this scenario capability to see what if this is gonna happen, so you can make a conscious decision. A nice example here is also something that happens quite frequently. The water level of the Rhine. It often goes dry or relatively dry, which means that the ships, and of course there's a lot of chemical companies on that, let's say Rhine closeness to do a lot of shipping via vessels and barges. And if it goes low the water level, this can't be done or they can't be fully loaded anymore, which has a major impact. So you might want to stock up upfront. And if you can run these different scenarios, you can check, Hey, what if I would stock up upfront? What does it cost me and what will be the impact if indeed this water level is gonna go down, versus what if I would not do it and a situation would occur? And then the conclusion could be quite simple saying, well, I'm more than happy to invest in this specific additional external warehouse in which I can store a little. It costs me a few thousand euros, but it's nothing compared to, let's say, loss that I have if the situation happens. So I'm willing to pay a little more to, let's say, mitigate the risk in full.

Tom Raftery: 9:19

Yeah. And where would you say decision delay hurts first? Was it customer service, working capital, production efficiency, margin, all of the above, something else entirely?

Robbert de Looff: 9:32

All of the above and in different places, depending on the specific focus of the company. So if you have your continuous, more upstream chemical units, they need to run 24 7. So you can't shut them down. Shutting them down is an, extreme cost, which is simply not desired. So for those, indeed, you see typically more the impact of the decision latency first and indeed the production process.'cause if you don't have the raw materials ready and you can't run at full capacity, your efficiency goes down. And of course, secondly, you'll start to see the need to push towards the customer is slightly different. However, if you swap it around and you look more into the, let's say, downstream of chemicals, so more indeed the specialty chemicals, think about companies that are selling semiconductor materials or paints, et cetera. For those, of course it's, it's slightly different because you can shut down units, et cetera. So I think here the decision lateness is probably coming at a higher cost if you want to maintain your service levels, 'cause you can swap around more easily, but you pay for it or you take the hit on your service.

Tom Raftery: 10:36

Sure. Sure. And what's one decision That you should never wait for the next planning cycle?

Robbert de Looff: 10:41

It's not really a specific type of decision. So it doesn't really a category of decision, but it's mostly on the impact. And that's also something very interesting that we see in this decision centric planning approach.'cause so far, indeed we spoke more about this process centric approach, but it's not so much about what kind of event or what kind of decision. It's about the impact of the decision. So what you want to do is you first want to do a quick assessment. What is the impact of a decision? And you obviously don't want to take every decision on the same second 'cause it's creating a lot of nervousness. If you say, today you need to produce A, but a day later you say, well actually it was B, and an hour later it's actually C again. Or it goes back to A, that just creates nervousness. So you need to have a very clear structure that, let's say creates how much the impact is of not making decision to see whether you want to act or whether you want to just make it go to the next cycle.

Tom Raftery: 11:35

So process centric versus decision centric. What does decision centric planning change in how people actually work?

Robbert de Looff: 11:44

It's a quite different mindset. So process centric is, is what we all know. It's all what we know from the textbooks. It's what we've been used to since, I don't know, in the nineties, somewhere eighties even. And it's very structured as well. So it's very easy to understand, easy to grasp. You do this on day one, day two, day three, day four. But if you change toward this decision mindset, it really comes with a different idea of how you run and how you make your decision. So one of those steps is indeed, as I mentioned before, classify whether you need to take a decision, yes or no. So is the event that is happening impactful enough to make a decision? Second thing is you don't have a structured approach of who should be in what meeting.'cause you don't have a structured moment of when the event hits. I mean, we all would've loved to be able to say when the Strait of Hormuz was gonna be closed or when it's gonna reopen again. But we cannot, I mean, it's a bit above our pay grade. We need to also have a, let's say, team organisation ready to, when we have a certain event that we can jump on it with the right people to actually get towards decision. That you have also the data available to get to a decision, the capability to run these simulations, to get to a decision. So it's not the standard cadence, it's really the capability, the organisation the people, and the, the solution as well to be able to get to the decision. So it's a different mindset. And I think here, I can even take that one step further because people often think about decisions like an event happens and I need to act. But in an ideal world, you're even ahead of it. You could have seen coming that the Strait of Hormuz was gonna close. I mean, not saying that you knew, but you could already say, well, it might happen. And this is the perfect example where you could say, well, what if it happens? Maybe I should already ask to get a few of those big boats with oil coming my way because I can hatch a little if it happens. And okay, I pay a few more, but it's worth it in the end. And, and this is indeed this, let's say, proactive look. So instead of predicting once it happens, you're gonna look ahead. What if this happens such that when it happens, you're ready and you have zero decision latency.

Tom Raftery: 13:56

Alright, and where does traditional S&OP still work and where does it start to break?

Robbert de Looff: 14:03

I would say it still works and I don't see it disappear that quickly. Also because decision centric planning on its own, in my opinion, you always still need to have that moment in which you look at your portfolio, you do the optimisation, who is the most margin based customer, who you want to fulfill, which contracts you can, cannot do, et cetera. So you will always need this structured backbone in which you do your optimisation. So I don't see S&OP the process centric approach disappearing in from a conceptual perspective. However it should be combined with this decision centric approach where indeed, while you're doing the optimisation or after you've done the optimisation, new events happen and those new events do need decision and not wait for a full month before you reoptimise them. So I don't think it disappears and how they exactly match and merge together is of course also very dependent on the organisation, how people want to run their business. And, and that will be very interesting times coming our way to, figure that out.

Tom Raftery: 15:02

And is the goal fewer meetings, faster meetings, or better decisions between meetings?

Robbert de Looff: 15:10

That's a good question. It, it's a bit of all I would say. The, the goal of meetings, of course is, I mean, my goal, at least always of meetings is to get the decision. Meeting to share information can be valuable at moments, but if you do a meeting, you want to get something out of it, you are bringing people together for a reason. And information can also be shared via different mediums, emails intranets whatever way you think of, but a lot of different ways. So I think meetings indeed need to get towards decisions. So meetings will stay around'cause you still wanna speak about things, you wanna evaluate things and get towards the conclusion. But a lot of pre-work can be done before going into a meeting. So more efficient meetings. as a result, faster meetings is at least the way I would position it if I would run a company.

Tom Raftery: 15:56

Okay. And what do leaders misunderstand when they hear realtime planning?

Robbert de Looff: 16:04

Yes. So as we spoke about indeed this cyclic S&OP, which is gonna stick around, I think this is a need, something that is often misunderstood. Realtime means a need that you want to, and actually also the event size. Realtime means that you do not jump on everything. It doesn't mean that you have realtime access to data changes. Realtime can only be relevant to specific situations. Let's call it like this. Realtime means that you have the data ready if you need to make a decision. So you get an event, an example, you get an event, and that event turns out not to be relevant. So it's a small strike in a specific part of a certain country because of which some of the demands might not be fulfilled. Okay? No big deal. You don't need to get all people involved. You don't need to global S&OP manager to come over or to, to get in a meeting to make a decision. I mean, in execution, you need to wiggle your way around it and next time you'll reoptimise to see how your distributions are looking at that moment in time. And that means realtime in a sense that you realtime make a decision. Do I want to act or do I not want to act? But it does not mean that realtime you're changing your plan and create a lot of nervousness. So realtime for me is on realtime getting information, realtime making decisions. But the decision could also be not to take a decision at that moment. And typically realtime is indeed more on the shorter term when you really want to see, hey, what is happening tomorrow? That's of course when you need to get that constant feedback loop from your ERP system to know what is being produced, how much do I have on stock? How much do I expect to be produced soon? This is indeed where you have that real, realtime planning.

Tom Raftery: 17:44

Okay, and can faster information sometimes create faster confusion?

Robbert de Looff: 18:19

Well, we always aim to have data coming from a single source. So one source of truth. That's always our aim. That's also what we, push for as much as possible because indeed, if you start to doubt whether the number you get in front of yourself is, right, that is of course very painful because it takes a lot of effort, a lot of time. And if you then afterwards need to send that to the group and you need to explain yourself, it's inefficient, what we just said we wanna avoid. So more data can lead to more confusion. But I think here it's important to select what data you wanna see. Today's world, there is a lot of data, like a lot, but not everyone needs to see the data to make a decision. So what you wanna do is you wanna evaluate with the customer in this case to see given a certain event, what data do you need to come towards the decision and how would your decision look like? And that amount of data that you need to make a decision that should be available the moment you make a decision. But all of the rest, it doesn't have to. Or maybe in the background if you want to deep dive, but not at the first glance.

Tom Raftery: 19:21

Okay, and practical perspective, let's say a vessel is delayed, energy prices spike, key raw material becomes constrained. What happens in a traditional setup and what should happen in a decision centric setup?

Robbert de Looff: 19:38

Yes. So in a traditional setup assuming their data is managed in a central way and all, let's say one single source of truth, which is already quite a big step, I'm assuming this is indeed the case, then you would have your S&OP cycle. And depending on where you are in that S&OP cycle, it might be that you just finished and you need to wait for another full two weeks before your next starts and your demand starts. Imagine that you have this spike in prices. It's gonna impact your production costs. It's even gonna impact demand potentially as well. So you really need to wait for a full two weeks, even a full four weeks.'cause the process also takes two weeks before people are gonna review the demands. Oh, hey, my prices are up. Oh, my demand is a little lower in that case. Production is gonna look at it as well and see, hmm, that's quite a high cost price. That means my margin shift.'cause now in Europe it's a bit less or a bit more, and in America it's a bit more, a bit less and well, et cetera. And you run these simulations, these scenarios as well. But it's, it's quite after the facts. So you only after four weeks decide, oh, with this, actually I might need to start producing somewhere in APAC instead of in Europe, which means four weeks long, you're producing at a cost which is higher versus what you could have done. So you probably already feel it coming. How that happens in decision centric planning. In decision centric planning, you get the event. So first of all, you check the event. What is the impact? Well, the impact is calculated. In this case, it's quite simple. I'm taking one example that you took hoping for the people in those companies. They do not all happen at the same moment. So a, a spike in the prices, spike in the prices is quite easy. You can see what the price increase is. The delta, you can also see how much it impacts your volume that you produce and how much you use because it's just a multiplication. You can see, hey, this increase in the, in the price is delta multiplied by the volume we do is one, two, 3 million, whatever number of euros that every day we lose. Or every week we lose whatever the granularities you want to see. And if that is above a certain threshold, you say, well this, we need to take action now 'cause this is important enough to make sure we resolve. So if you resolve indeed, you make sure that with this, you see who, okay, who's owning this decision, decision centric planning approach. And that person then needs to make the call. Who should be involved in making decision? What data should be present to get towards the decision? And what simulation should I run upfront already to get towards the decision? That person can make sure he or she executes, of course these kind of activities and has a prepared set ready and says, well given my base plan that I have, given this event that happens, we have different options we're gonna produce in APAC as you just gave as an example, we can stop production because it's simply not profitable anymore. Or we can buy it somewhere in the market.'cause we still have some, let's say partner, which is having a lot of stock, which is still at the lower price. So you get three kind of options and then you go to the meeting with those right stakeholders involved. They say, well, we go for option A. Or maybe they ask, can't we do B with a bit of this? You try it, you get an option four, so to say. Doesn't We go for option two and within one day, let's say you have the meeting, you have the decision and you can stop producing in the expensive plant, saving you that number that you saw from the initial check.

Tom Raftery: 23:02

Fair enough. And is this something that AI can help with as well? And if so, where should it help and where should humans absolutely stay in control?

Robbert de Looff: 23:13

There's different ways and different places where AI can help. And I truly believe in a synergy between AI and humans, at least at this point in time. Maybe in future it's different, but for now we need to collaborate together. One is of course in your navigation within the application, 'cause you still have one person that needs to do the checks of what data needs to be available, what kind of options do we have ready? And that person can work more efficiently using AI. Using AI in a sense of instead of clicking through the application yourself, you can speak in natural language like, Hey, can you try me a scenario in which producing more from here? What if this, what if that? And you have the AI create the simulation in the background. Using the application as it always existed before, of course. But using that in a natural language way and also visualising it easily to, or you, so instead of you searching for where to click to see it, you now just get it opened up. You get the pivot with the KPIs to say, yes, I want to go for option A. So that's one. AI can help in making it more efficient to prepare for these meetings. But a second as well is to get in the events. Because events, there's many kind of forms of events and as you might think about it yourself as well, it's not so easy to get a specific data source of events.'Cause who knows that there is gonna be a strike in France tomorrow. I mean, that's in the news. Yes. So if you read the news, but you can't read the news for all companies or all countries close by. Water level of the Rhine is another one. It's not something that resides in your ERP system, of course. This is where AI can help as well, where you can indeed look at, hey, what is now relevant to my business and in my area of my, let's say producing sites or distribution centres. What kind of topics do I want to get flagged? Or trends. Trends on, for example, electricity prices or trends on oil prices. Hey, what if it goes down? Flag me if it goes below a certain threshold. So that's where AI can help as well and need more of this, let's say web search. So having all of the information and how can I get these events which are relevant out.

Tom Raftery: 25:09

All right. And what improves first, when companies get this right? Is it cost, service, inventory, confidence, something else entirely?

Robbert de Looff: 25:21

So I wanted to say confidence. Confidence is, a very interesting one. The others of course are part of, it as well, but confidence is a key one. Structure, stability, confidence. And here indeed you do not always want to reward the person who is extinguishing the fire. You also want to look at the person who actually make sure that the fire doesn't start in the first place. And this is something I've seen in our company as well. By doing this decision centric planning approach, they create stability. They just know when there's an event, we know what to do, typically even upfront because we thought about it already before and we already know what the resolution is. So instead of creating a massive stress when such an event happens, it's just normal. It's like, well, we know we've seen it, we've done it. Now we just need to execute it. So confidence, I think is a, is a key one. And obviously the others as well. I mean, with these kind of better decisions and faster decisions, you also do see a, a benefit and increase in service levels and very important for that specific company. I was, I was thinking it's Evonik actually because we have a case study on them as well. They are in the middle in chemicals where they get this raw materials from a continuous provider and they need to deliver lot of their materials as well towards continuous providers. So as you can imagine, they all cannot stop working because if one stops working, the other one has a massive problem in a sense of they need to shut down or scale down their units. So for them, there's also benefits of course, service level because now they are proactive and they can actually make sure that they can deal with these unforeseen activities faster to make sure that their silver level set up. And that all boils down to bottom line as well.

Tom Raftery: 27:01

Sure, of course. Yeah. Yeah, yeah, yeah. And in in large planning transformations, what's harder? Is it improving the numbers or getting teams to trust the new way of working?

Robbert de Looff: 27:13

The latter. I would say the latter is very tough, and it depends on the size of the company as well. You see here indeed, that typically the more lean teams are a bit faster in doing this kind of adoption. But of course the, the more people are involved and also the more sites are involved, the more tough the decision making is getting.'Cause everyone needs to be involved. Everyone needs to have an opinion and everyone needs to of course say yes in the end. so I would definitely say the latter. The people are typically indeed the more challenging part. But you do need the foundation again. You do need the solution in which you can track it. You also do need the process to track it.

Tom Raftery: 27:48

And how, does the planner's role change when routine manual work is automated?

Robbert de Looff: 27:55

Yes, it does change quite significantly.'cause before you used to be indeed a planner, you're doing your plan. And now you're not really a planner anymore. You are some kind of, yeah, creative scenario creator, which is thinking about what if this, what if that, what could be the, so you're really changing your, let's say, way of working from going to, these are the facts and I just need to make sure A matches with B. And now you're going to, well, maybe we should think about what A could be or what B could be. So it's, it's really a change in, I would say, character as well, but also skillset. You still need of course a lot of the skills which the let's say traditional process centric planners needed.'cause you do need to know the business. You do need to know how planning works. You do need to know supply chain. But there's a lot more coming their way because indeed the part is automated. And indeed it goes more into this creative mindset or open mindset of what could happen.

Tom Raftery: 28:51

What tends to be harder than executives expect when they're moving to this model?

Robbert de Looff: 28:57

Yes. And I actually already came back in the beginning a little when I said I assume data to be ready. This is often a wrong assumption in a sense that often data is, is a big challenge. There's all of data available but not cleaned in the right way, not documented properly. Parts are missing and, and this is a key topic. A key topic is data. And that's really the first part the journey. And also a part which is often forgotten because if you do planning with data, which is slightly off, so think about a rate which you set at a hundred, but you can actually, in reality only produce 80, you're gonna be impacted on your service level.'cause each day you're gonna be lower on what you actually produce versus what you've planned for. So, getting the data ready without even having a solution in place, just using what you're doing today is already improving your planning results and as a result, your service level, working capital and your bottom line. So, data is one of the, key ones, the starting ones. But I think afterwards what is also typically very challenging is the change management part. The adoption part.'Cause often in the organisation, people who've been doing the work for a lot of years already. They know it inside and out. They also know how their Excels work. They've created the Excels themselves and no one really knows what does what. Sometimes you even have a heritage, Excel of 20 years ago, and there's a number in there. And when you ask what does the number do, they say, I don't know, but I don't wanna touch it.'cause if I do, it breaks. And that's also a very tough part.'cause getting such a person transitioned into indeed another way of working is something that, does need very strong support or need a project sponsor from an executive to say, look, this is the direction we're going. We're gonna do it all together. We're gonna help you to make sure that they're all on board.

Tom Raftery: 30:40

And What kinds of decisions should companies rehearse before disruption hits?

Robbert de Looff: 30:46

Like running, what kind of simulations to run, what kind of, what ifs to do? Yes, yes, yes. I think here it's mostly about the bigger ones. So what you do not really, get a lot of value from is doing the small change within a specific factory. So you can do a small change like, Hey, what if this specific raw material or packaging material is not fully ready on this specific site yet? Well, it's an important what if, but it's on a lower, smaller scale. It's only impacting that specific plant. It's only impacting a small part of the volume and small part of the revenue. For those, I would say still interesting, but less the big leap forward in savings. It's more the, the bigger decisions, the bigger decisions that impact multiple teams that impact multiple horizons potentially, or which have a long lasting effect. Think about indeed the energy prices spiking because of the war, Ukraine, Russia. That was a very interesting and very key one. Think about indeed the Middle East now, or the, Red Sea conflict as well, 'cause of which there's lower supply, but also because of which sometimes the lead times change because they need to take a different route. And those you see directly as a major impact.'cause you start to shift, your production volumes may be regionally or you start to shift your demands regionally. So you really see this global impact on a larger scale. would say that's where you can take the, the biggest leap forward.

Tom Raftery: 32:07

Okay. And over the next three to five years, what's gonna change most in supply chain planning?

Robbert de Looff: 32:12

A lot is gonna change I think. If you look at the speed of technology, it's impressive. It's impressive. And maybe even a little scary 'cause if you think about Open AI launching their ChatGPT, that that's when the world really started talking about AI. And of course it was there before, don't get me wrong. That's when it really boomed. And in those few years that we had since then, it became so normal to use it. And a lot of applications of how to apply it in use cases are already available. And I feel like this is not a linear growth. I feel like we're in this kind of exponential curve, which you always have into the, let's say the different historic moment. I feel like we're at the beginning. I'm very curious to see if this exponential curve is gonna slow down soon, but I don't think so. I think it's just gonna keep going for quite a while. So, what is the biggest change is, it's tough to answer because there's so much unknown today. But I think that it's gonna move more towards this autonomous kind of planning. Autonomous kind of planning meaning indeed that these recurrent decisions that need to be made are just made automatically. You do not really need to be involved.'cause it does it very well, it might even do it better than the person who used to do it manually. I think what is still gonna stick is the relationship building. In the end, decisions are very often connected to a certain relationship you have as well. You can't just have two AIs talk to each other to say, yes, we go for A, so I think here indeed, you still need the AI to do your work, to autonomously plan, but to really get towards a conclusion, and decision, with your partner, with your customer, with your supplier, I think you need to use that information, that autonomous plan that's created to get towards an agreement.

Tom Raftery: 33:51

Okay, it is time now, Robbert, for the lightning round. So I'll ask some questions and you've got a one sentence answer, okay?

Robbert de Looff: 34:00

Ooh, yes.

Tom Raftery: 34:01

Okay. What's best speed or certainty?

Robbert de Looff: 34:05

Certainty, because you can be speedy, but if you're wrong, you're still wrong.

Tom Raftery: 34:10

You're wrong more quickly. Okay. Lowest cost or lowest risk?

Robbert de Looff: 34:15

Lowest cost because if it's low cost, you can try it again.

Tom Raftery: 34:18

No good. What delay hurts most?

Robbert de Looff: 34:22

Delay in decision making.

Tom Raftery: 34:24

Okay. What should never be automated?

Robbert de Looff: 34:29

Relationships, never, personal connections.

Tom Raftery: 34:31

Nice. What's the biggest planning myth?

Robbert de Looff: 34:35

I'll get a solution for it and it will work.

Tom Raftery: 34:37

Okay. And what's one decision leaders avoid?

Robbert de Looff: 34:42

Ooo that's a tough one. I think they're typically a bit risk averse. And taking that leap I think is gonna help you move forward. Fail fast.

Tom Raftery: 34:51

And no longer lightning round. If a leadership team wants to test whether they have a decision lag problem, what's the first question they should ask themselves?

Robbert de Looff: 35:01

So I think, first of all, they need to check where a bottleneck is. Is the bottleneck in the people not connecting fast enough? Is the bottleneck in not having the data ready and visible? Is the bottleneck in not having the solution to make different scenarios? Or is the problem indeed in not having a process to get the decision towards execution? Because it could be that you can make a decision very fast, that if you don't get it to execution well, it was a fun decision, but not really gonna do anything. So, first step is really finding what is the bottleneck?'cause that's the one that needs to be resolved first. Afterwards, you can see how indeed you can improve.

Tom Raftery: 35:34

We're coming towards the end of the podcast now Robbert, is there any question I didn't ask that you wish I did or any aspect of this we haven't touched on that you think it's important for our people to think about?

Robbert de Looff: 35:45

No, I think we covered it all nicely. I really like the fact I made about this confidence booster. So really want to emphasise that, that it's, it's not always about the results. The results do follow, of course, but it's, it's really neat about also the feeling that it brings and the efficiency that it brings and the, the calmness that it brings. And the second part, what, what I really liked in need is this connection of AI. Of course, AI can bring a lot, but it doesn't mean it's gonna push away personal relations that you have. I think they're still key. They'll remain key, at least for a very long time, at least I hope, as long as I'm still around in here. So no, I, I like the fact that we touched upon the topics, which are, let's say, not specific to the technology, but also a little bit more the personal side of it.

Tom Raftery: 36:24

Brilliant. Alright, super. Robbert, that's been fascinating. If people would like to learn more about yourself or any of the things we discussed on the podcast today, where would you have me direct them?

Robbert de Looff: 36:34

You can always find me on LinkedIn. So with my name Robbert de Looff, of course. If you would like to know more about OMP and our Unison Planning Solution, of course you can also visit our webpage omp.com.

Tom Raftery: 36:45

Great, Robbert, that's been really interesting. Thanks a million for coming on the podcast today.

Robbert de Looff: 36:49

Perfect. Thank you. Thanks for having me and a great conversation.