Developing our Design to Value approach as pharmaceutical plant design consultants

In 2012, Bryden Wood and GlaxoSmithKline started working together to try and come up with a different approach to pharmaceutical capital projects. We started from the premise that while a great deal of effort was expended by a lot of expert people within both pharmaceutical and engineering companies, and despite the significant sums being spent, capital projects often failed to meet business requirements. Why was this? And how could we do better in our approach to pharmaceutical facility design?

We developed a particular set of tools and an approach that over time became what we now refer to as Design to Value. As we have evolved this approach over the years, we’ve extended it to include many more pharmaceutical engineering projects, as well as a wide range of other sectors, from heavy industry to prisons.

Part of our role is to act as pharmaceutical plant design consultants. What we aim to do is make sure that projects are well aligned with business requirements; to define and design future assets that translate a company’s strategic objectives into tangible manufacturing facilities – and provide added value. The key point here is that we put a great deal of effort into defining the business needs. That is, after all, why we're doing the project.

We also invest effort into constantly assessing how well our design solutions match those needs. This has often resulted in solutions that require no capital expenditure at all, because we find different ways of meeting the requirements.

The Design to Value approach relies fundamentally on close collaboration between a wider range of experts than is seen in conventional projects, particularly including business functions. These experts are actively involved in the definition and search for solutions on an equal footing with the engineering and architectural people who would normally be considered part of the project team.

We also use an iterative approach to pharmaceutical facility design, recognising that good solutions emerge from exploration of all the options. So our chances of finding the optimum solution are increased when compared to more linear design approaches.

There are four principles that enable us to operate in this way:

  • A common language allows people with diverse areas of expertise, including those not normally involved in design and construction, to participate fully in design decisions. This is much more effective than a conventional approach where you might produce engineering drawings or documentation and then hand them to a user group who will then take additional time to review and pass comment. And, of course, if people are fully involved in generating solutions, they're more likely to adopt them.
  • Visual information is vital in communicating complex issues. A generally accepted principle in systems thinking is that diagrams are required to convey complex issues in a way that text generally can't.
  • We also use modeling and data visualisation, to ensure that decisions are based on facts and data, rather than opinion and hunches. Transparency of data sources and decision-making is important, because it gives people confidence in the analysis and the proposed solutions.
  • Knowledge capture allows reuse of learning from project to project, promotes standardisation, and allows us to trace decisions through the life cycle of a project.

We will see how Chip Thinking® plays an important role in applying these principles.

What are Chips?

Chips are the smallest meaningful part of the processes that make up a supply chain. Chips are sets of interacting or interdependent components, plus all the data that goes into them. They bring to life all data and stakeholder knowledge in one place, using a common language, so that everyone can understand every aspect.

Chips become the building blocks for our design. They also become the building blocks for conversations between the people dealing with business strategy and those people dealing with design and implementation of the pharmaceutical engineering project.

We start off with the analysis of the business problem and then generate design elements. Chips will often appear as 3D models, but that's not what they are. They are the embodiment of a system. The core purpose of the project drives the analysis that informs the selection of Chips.

Defining Chips for the project

At a point in each project, we go through the process of defining Chips for that project, in order to develop a Chip library. This can start with a reference design, or a blank piece of paper. Or we might start with a similar type of project for which we've generated Chips in the past.

Chips can be made at different levels. A chip is not a module or a piece of equipment, it's a part of a chain. Depending on the problem you're trying to address, you can set the level of the size of the Chip to the level that is appropriate to your analysis. A chip can, therefore, be a whole factory, a building, a production line, or a single machine within a production line. These different levels will often result in a hierarchy of Chips, so that big Chips can later be broken into small Chips. They're used in different ways at different stages of a project, or in different ways by different users within a project.

The data structures that we build around Chips allow data to be aggregated between different levels.

We can associate any type of data with a Chip. Conventional engineering data is perhaps the most obvious, but we also include data like staffing levels, containment requirements, power consumption, or even the level of design uncertainty. This allows for visualisation of different issues within a project.

It's important that when we define chips, we don't leave gaps. We aim to capture everything: the building, the equipment, operations, software, hazards, quality requirements, whatever is important in that system. At the point of inquiry, we are trying to get that information loaded into that Chip.

When we are creating plans, we like to work in three dimensions from the very early part of the project. Chips help speed up the process, because we can move chunks of the design around very quickly and keep data associated with them. We can do things like automate routing of certain utilities, based on the properties of the Chips. And that enables us to try lots of different options in a short amount of time. As mentioned, this is a key element of our Design to Value approach, as it enables us to get closer to an optimum solution.

Unfortunately, we sometimes find ourselves in the conventional engineering situation of having to limit the number of options that we look at because we haven't got the time or the resources to look at too many. Clearly, it is fine to narrow down a set of options on a rational basis. But if it's simply that we don't have time, we are at risk of staying away from the optimum solution. That means more cost and more time which, ultimately, is going to affect the patient who is at the end of that supply chain. As a result, these are particularly important factors to consider in the design and construction of pharmaceutical facilities.

In summary, Chips allow us many different perspectives on parts of a project. And crucially, they provide the common language for communication between all the different people involved in executing that project.

Improving pharmaceutical facility design with Chips, generative design and automation

We are now using chips as the basis of generative designs to allow us to automate the design process. In this context, generative design means using computers to assemble designs based on a set of rules, components and input parameters. We generate very large numbers of options for a particular design requirement, and then get humans back in to look at those critically.

We can generate many more options using automation. Sometimes we find something which human expertise has not spotted. Generative design can create unexpected options with high value.

We are now working on making Chips available to very large numbers of users through a web interface, to use generative design for pharmaceutical plants. We are also associating process simulation data with those Chips, so that we can look at throughput, and use throughput to define what Chips we need.

The Chip as future enabler

We are often asked what ‘Chip’ stands for. It doesn't stand for anything. It's based on the idea of a microchip within a computer. The microchip is the bit that does the hard work. It's the intelligent bit and the bit that adds value.

Chips are enabling tools that do a lot of work and add a lot of value. Within our wider Design to Value approach and methodology, they provide a common language for all the people involved in a project, they enable collaboration and the rapid development and testing of multiple ideas. We've seen the benefits they deliver.

Using the same concept to enable advanced and automated design will only multiply those benefits.

 

To learn more about our Design to Value approach, sign up for our monthly newsletter here: http://bit.ly/BWNewsUpdates