Standardisation is not always cheaper
The competition is intense. Consumers are becoming more demanding and companies are looking to respond optimally to demand. The result: increasing product differentiation and a more complex, and therefore more expensive, logistics chain. Is a platform approach the solution? Together with professors Robert Boute and Behzad Samii, Vlerick researcher Maud Van den Broeke has developed a model that companies can use to calculate how many and which different platforms they should preferably develop for which (end) products. The model is being validated in collaboration with Barco’s Healthcare division.
The general public probably best understands the platform concept from the example of the automotive sector. “The idea behind it is simple,” explains Maud. “While each model has its own look, cars of the same make have parts in common such as the chassis, power trains and electronics. The more you can standardise, the less complex the supply chain becomes. And you can then develop the versions you need from the standardised platforms.”
Barco Healthcare is also using the platform approach. The division develops and produces medical screens – monitors that are used for radiology, mammography, operations and dentistry, among other things. “In just the same way that a car make has various models and variations – big and small, slow and fast – we have big and small screens, with high and low resolution, colour and greyscale monitors etc.” explains Kristof Deneire, Supply Chain Manager at Barco Healthcare. “When we develop these different screens, we try to reuse as much of the electronics as possible. The printed circuit board assemblies – circuit boards with electronics components – are our most important platforms. You might compare them with the engines in cars. The same engine can be fitted in different models, albeit within defined limitations. An engine that is suitable for a small city car is hardly likely to be fitted in a four-wheel drive. But using the engine from that SUV would be unnecessarily expensive in the same city car. The question for us is how many types of printed circuit board assemblies or electrical diagrams do we have to develop to yield the benefits of reuse?”
Total logistics costs
Just how far does standardisation need to go? The team has developed a model that evaluates different product platform scenarios, ranging from full standardisation – all end products derived from the same platform – through to no standardisation – a separate platform for each end product – and every possibility in between. The total logistics costs are calculated for each product platform scenario:
1. development costs – costs associated with the development of the various platforms and the resulting end products;
2. purchasing and ordering costs – costs for the purchase and ordering of platforms and additional components;
3. stock costs – costs associated with working and buffer stocks, including costs incurred due to stock deficiencies, both for platforms and for additional components and
4. transformation costs – costs associated with actually deriving end products from the platforms.
These costs can also be grouped into costs which are related to the platforms themselves (1, 2 and 3) on the one hand, and the costs needed to adapt the platforms to the end products (1, 2, 3 and 4) on the other hand.
Maud summarises their approach as follows: “We created an inventory of all the choices that need to be made in a platform scenario like this, and looked at what the impact is on every aspect of the logistics chain. This impact is then quantified, allowing you to compare different scenarios objectively.”
The optimum scenario is the one where the net present value of the total logistics chain, calculated over the expected life cycle of the platforms in question, is the lowest. And that is not necessarily the scenario where every product is derived from the same platform. “The platform concept can be seen as the pinnacle of logistics efficiency – as if standardisation is by definition cheaper. Which is of course not the case,” says Robert. “The more standardised the platform, the more modifications and thus additional costs you incur if you want to offer the same product differentiation.”
“Our model clearly demonstrates that you need to consider platform costs against adaptation costs,” explains Maud. “Higher platform costs generally mean lower adaptation costs, and vice versa.”
This is influenced by (1) the number of platforms being used and (2) the extent to which these platforms are under or over-dimensioned, which is further determined by the decision as to which platforms are developed and which products are derived from which platforms. “If under-dimensioned platforms are used or there are a few platforms for a lot of end products, the platform costs will be lower but the adaptation costs will be higher because the end products will require more modifications and additional components. Conversely, more platforms and over-dimensioned platforms will mean higher platform costs, but then lower adaptation costs at the same time,” she affirms.
The model has been applied to the concrete platform scenarios of Barco Healthcare. Has the company changed its approach on the basis of the results? “The simulations confirmed in the first instance that our approach is good, and that we have got it more or less right in terms of the number of platforms,” says Kristof. “The analysis demonstrated unequivocally that by investing a little more time in developing a platform, you can save time on the manufacture of the end product.”
In the background to the simulation, the drivers for bringing new products to the market were also investigated. “In our industry, the main driver seems to be technological change. With technology evolving so quickly nowadays, it is best to get your new products to market as quickly as possible. This model can help you do that. And when we considered basing a new product line on a combination of two platforms, the model advised us not to, as the simulations indicated that it would not make financial sense.”
And what is the biggest benefit of this research? Kristof has no doubt about it: “It has succeeded in translating a discussion which may otherwise have remained philosophical and subjective into euros, allowing us to reach better substantiated, objective decisions.”
The research and development of the model are in the context of a European research project, Varies (variability in safety critical embedded systems), which forms part of the Artemis research programme, supported by the IWT. Twenty-three partners from seven countries took part in the Varies project, among them Barco and Vlerick Business School.
Source: The paper ‘Evaluation of product-platform decisions based on total supply chain costs’ appeared in the International Journal of Production Research, 53:18, 5545-5563. It includes an extensive discussion of the scientific model and its application in the concrete example of Barco Healthcare, including sensitivity analysis.
About the authors
Maud Van den Broeke is a graduate researcher at the Vlerick Business School. Robert Boute, supervisor of Maud’s PhD, is Associate Professor in Operations Management, and Behzad Samii is Associate Professor in Operations and Supply Chain Management. Both are associates of the Vlerick Business School.