As the number of clinical trials increase in both complexity and scope, the pressure to develop products quickly in a cost effective manner has never been greater. With things moving much faster in the development space, the ability to handle the demand of supplies and make accurate forecasts is critical to the success of a clinical trial.
In this interview, CTA editor Henry Kerali speaks to Johan Verboven, Senior Project Manager, Business Process Design at GlaxoSmithKline Vaccines. With years of experience working within clinical trial supply management, Verboven knows how to navigate what is an increasingly complex supply chain. Below, Verboven discusses how SAP software can be used effectively to optimise supply chain processes.
Clinical Trials Arena: What are the advantages of using SAP software, and how does it help in terms of managing your supply chain?
Johan Verboven: It’s extremely useful because it’s a planning tool. For GSK, we are an SAP based company and we already have our data in SAP. I’d say using SAP to plan that data while also using APO (Advanced Planner and Optimizer) is a logical solution.
In the past, we used Excel along with VB (Visual Basic) to schedule our activities. Naturally, Excel has a lot of limitations and isn’t the most effective tool for forward planning. So there was a lot of work for our planner to schedule everything accordingly. With Excel, we had a lot of peaks and troughs in our capacity and this is something that is automated in the planning tool of SAP. When you run the planning – and this is something that we do twice a week – SAP takes care of all that. It prioritises based on the shipment date and then will fill capacity where possible, and all it involves is the push of the button and the system takes care of everything.
Additionally, our planner function is moving more toward to something that is analytically based where the planner will manage capacity and notify management what may occur in the future. In the past, we used Excel as a scheduling tool while now we’ve added a forecasting based workload estimation (in SAP it’s called ‘Rough Capacity Planning’) and put in place milestone planning. These two additional tools enable us to have a more pro-active planning approach.
The timelines we agree with our clients are also modifiable, so if for a certain reason one project needs to go faster that is perfectly possible, whereas in Excel, this is not something you can easily manage. Before, this would have been done manually and now you have the big advantage having all of the data made available with the simple push of a button. And it’s all integrated and changed in real time, so if tomorrow something changes in our forecasting, you will immediately see it (or the day after) in our rough capacity planning. So you gain a lot of manual data entry that you avoid.
CTA: Using this forecasting tool has many advantages over an Excel spreadsheet, it seems. But what are the some of the downsides?
JV: One particular downside is that your planning run is a ‘black box’, and what I mean by that is when you push the button, SAP will do its work and you get an outcome. But it’s very hard to determine why the system plans certain things accordingly. This is something in our implementation phase we tested thoroughly. Once you go into production, you don’t really control it anymore, so it really requires a lot of testing to ensure that all the scenarios you could encounter are tested and that the system functions the way you want it to. The fact that it’s a black box is in a way a downside; it’s a very complicated tool and it’s not really easy to assess and ask why it didn’t it schedule accordingly.
Another downside is you need to maintain master data. The system bases itself on data, for instance, a labelling run of 2000 vaccines will take ‘X’ amount of hours – that data you need to put in the system and based on that data the system will schedule. So the quality of the data you put into the system is very important because if you input bad data, your output will be of very low quality. You definitely need to take into account that this master data needs to be maintained as well – any changes in processes need to be modified. In a way, it makes sense, but it’s additional work you need to take into account. This is something you don’t have when you do everything manually, which is a big difference.
CTA: In your experience, what have been some of the best ways of implementing SAP?
JV: Looking at planning more broadly, there are two philosophies – infinite planning and bottle neck planning. Bottle neck planning means you schedule all your activities based on your most constraining activity. Infinite planning, on the other hand, means your planning tool will take into account every single activity, resource, production line, etc. We opted for the latter, which is much more difficult than bottle neck planning. With bottle neck planning, you take into account one team, the most constraining and assume that all other teams will follow. If you produce something you schedule – your production, for example – but you don’t schedule your QA resources because they assume they follow what you produce. In the Clinical Supply Chain we don’t; we take into account all involved teams and that makes your planning tool much more complicated. This is something you need to consider upfront. When I say our planning tool is a black box it stems from the fact we opted for infinite planning.
If you want my advice, if I were to make the same choice again I would lean toward bottle neck planning, largely because it’s much simpler, it’s easier to implement, and it’s easier to modify. While it has downsides, you make the assumption that things will follow, but that’s not necessarily the case.
Another thing we learnt is that your process needs to be mature, prior to implementing a tool. If you don’t have a mature process, you may modify your planning tool unnecessarily, so you really need to know what to plan, and the same goes for your master data – it’s needs to be of good quality and updateable.
CTA: What would you say is the key to developing an effective simulation strategy?
JV: A lot depends on our rough capacity planning and the aim of that tool is show whether our demand matches capacity or not. The outcome is valued as a percentage whereby if the value is under 100 percent it would be green and it’s over 100 percent, it would be red – red meaning you have a higher demand than your capacity. The purpose here is to have a proactive approach to planning whereas in the past, we were very reactive. For example, from today if we know in advance in August we’re going to have a peak in workload, we try to do what we can to mitigate that by doing things earlier or later. But if that’s not possible, we can hire people, or bring in people from other departments, so having the ability to see what’s upcoming is a huge advantage. While in the past, we had a manual planning tool that didn’t give us this kind of visibility. We resolved the problems when they arrived. Here it’s a different approach and it’s something we try to implement with the system, but it just shows data and it’s up to our planner to analyse and to escalate to management and highlight we will have a problem in August, for example.
*Johan Verboven is the Senior Project Manager, Business Process Design at GlaxoSmithKline Vaccines