De ci, de là, cahin-caha, va chemine, va trottine

Grosso modo :

Toute décision de conception d'un système a un impact sur la performance future du système et son environnement : profit, meilleur service, sécurité et bien-être, plus rarement efficacité écologique.

Le découpage conceptuel d’un système en sous-systèmes, qui permet de passer d’un problème complexe à une liste de sous problèmes spécialisés plus simples à manager, favorise des prises de décisions locales  optimales sous réserve que rien ne change par ailleurs.

Mais dans un tel processus, toute tentative d’optimisation de la conception du système à un niveau global ne peut être réalisée qu’en appliquant des améliorations incrémentales ponctuelles, puis en observant leurs impacts sur le système en terme d’interactions entre composants et de changement de comportement global du système.

On peut résumer cette situation en disant que le système global émerge à travers des itérations de conception et d’essais plutôt qu’il n’est conçu activement, et qu’en général jusqu’à présent les décideurs ont toujours limité sévèrement toutes itérations, ne serait-ce que pour des questions de compétition qui induisent implicitement d’acquérir un avantage concurrentiel juste nécessaire pour pouvoir faire face à n’import quelle évolution future du marché.

Finalement, en matière de gestion des ressources, alors qu’au cours du cycle extraction, utilisation, gestion des déchets, nous ne payons que le terrain, les coûts du travail et de l’énergie et le moins possible à la planète pour les dégradations, la conception d’un système de gestion des déchets au niveau global est d’autant plus difficile que cette gestion est indépendante du système qui  génère les déchets, et que chaque producteur de déchets peut choisir d'accepter le coût supplémentaire d’une taxe pour autant que leurs concurrents suivent le même chemin

En contrepartie de ces abstractions étranges sur la conception des système, la conférence de Pierre RABHI au Corum de Montpellier relatée en 2 partie par Montpellier Journal :

http://www.montpellier-journal.fr/2009/09/nous-sommes-au-...

http://www.montpellier-journal.fr/2009/09/%e2%80%9cune-fa...

Steve Evans, 8th September, 2009

Design is everywhere, deciding how things look, how they work and how systems run. But effective ecodesign is difficult when decisions are made in isolation, says Steve Evans

The greatest challenge is the industrial system itself

Everyone does design. Some of us have qualifications in it. But others do it by deciding what should be in an insurance policy, how polite the waiter should be, when the bus arrives or the shop opens, or even how many bins are to be collected.

All these decisions are design decisions shaping the future performance of the system, including its environmental performance.

We typically design something with a particular intention: for profit, or to deliver the maximum service or welfare. We rarely design for eco-efficiency. Achieving this requires that environmental improvements in one part of the system are not immediately lost elsewhere.

A good ecodesigner should be worried about how their decisions affect other connected parts of the system. But individuals are usually limited in the scope of their influence. For example, the insurance designer will find it difficult to change the shape of the health system or criminal justice system, so will naturally design their part of the system assuming that other parts remain stable. Most designers are busy optimising sub-systems, while no designer is responsible for the whole.

Ecodesign has therefore been limited to incremental improvements. Waste is one of the most challenging outcomes of a poorly designed industrial system. Indeed the industrial system as we currently recognise it is emergent rather than actively designed.

Competition is unlikely to change this as each actor cannot guarantee that other actors will co-operate. Yet guaranteed co-operation is needed before investment can be made in new business models, such as take-back and leasing.

Many excellent writings and teachings emphasise ‘system-level change’ or whole system design. This concept contradicts many of our professional instincts.

Instead of starting with a tough problem and reducing it to sub-problems allocated to subject experts, resulting in solutions we expect: more technology and incremental performance improvements, we embrace the whole problem and look for useful interactions between the components.

This is obviously difficult to accomplish, and possibly mad when described at such an abstract level, but an example may help:

• We can make Internal Combustion Engine (ICE) cars that do 50 mpg. Indeed, 70 mpg is becoming the benchmark for new car launches. Hybrids are more complex and perform about the same as the better diesel cars. Electric cars offer emissions at the power-station, with the G-Wiz delivering the CO2 equivalence of about 150 mpg but current hydrogen fuel cell cars are no better than the average vehicles of today.
  None of these deliver the radical changes in performance that we urgently need.
  
• Amory Lovins, of the Rocky Mountain Institute, addressed the technology challenge 15 years ago by proposing to design a car around the hydrogen fuel cell rather than making a fuel cell behave like an ICE. By investing in making the car lighter and more aerodynamic, the fuel cell would be smaller and cheaper and the resulting fuel savings would pay for any additional vehicle cost. Lovins recognised that this would only work if the vehicle was leased and not sold.  Yet these cars never made it to the market.
  
• That thought troubled Hugo Spowers, one-time owner of Prowess Racing, a motorsport business. He spent years developing a new governance structure for his business, now called Riversimple, which makes all stakeholders equal partners, removing the dominance of founders and investors.
  
  Concerned that a minnow could not compete with the big companies with their R&D budgets, Spowers proposed to make the car design open source, gifting the design to a foundation and encouraging others to develop it further and faster than the big fish could. Finally, to reduce the financial barriers to car-making, Spowers chose to design the vehicle to be profitable when built in factories producing 3,000-5,000 cars per year.
  

Many of these solutions run contrary to current wisdom. It is hard to imagine a current car company making their designs open source for example. So far Riversimple’s whole system thinking has led to multimillion-pound investments and a demonstration two-seat urban vehicle that does the CO2 equivalence of about 300 mpg.

• For buildings, it seems obvious to invest up-front to reduce energy consumption through the life of the building. But at what, and at whose, cost? Developers and first owners may feel that they are unlikely to get their investment back and the calculations have to assume that other variables remain unaltered to make them possible.
  
  So we plot curves of incremental levels of insulation versus the energy saved. It is not at all obvious that spending more could eventually lead to such sufficient levels of insulation that heating and cooling systems would not be needed. Removing traditional central heating systems saves a lot of money in house building, making the extra spend on insulation feasible, but this is not built into the traditional cost versus insulation calculation.

These examples emphasise energy during the use phase of the life of a product. This simplifies the whole system design problem and encourages solutions where companies maintain their relationship with the product through its useful life.

• Waste is more challenging. There is no system more dysfunctional than that of extracting and using resources and managing waste. The language itself emphasises separation.
  
  Managing waste ignores other parts of the system.  If we want to design out waste, we have to start at the level of the whole system.
  
  For example, we do not pay the planet when it provides us with value, we only pay the direct land, labour and energy costs. The end-of-life equivalent: landfill tax, is beginning to bite.
  
  But even this is failing to translate into broad responses, as each waste generator can choose to accept the extra cost as long as their competitors follow the same path.
  

The design of economic instruments that encourage innovation and competition must be an ecodesign priority and individual producer responsibility is an excellent start. The greatest challenge is the industrial system itself.

• Waste producers must learn how to reduce waste, while the experts, the waste management companies, are hardly incentivised to go to their customers and show them how to reduce their custom. The system acts against innovation.
  
• Ecodesigners should be directing their efforts to understanding what the current best in class performance is for all products and systems, and how near (or far) the majority of products and systems are from this.

Then they must demand best in class products and manufacturing practices from suppliers. This works for government procurement and, through legislation, for consumer products and systems.

They would ensure that knowledge of the full energy and resource ‘shadow’ for all products and services are available to producers and consumers and would support massive re-education of the existing workforce.

Finally, they would be very busy designing systems that support and reward significant reductions in energy and resource use, and they would facilitate industry co-operation to deliver whole system-level change.

Ecodesigners have to be willing to embrace the mess of the whole system.

They must actively seek to co-operate with others, and immerse themselves in the reality of the detail.

If we are to transform from a linear model of extraction, use and loss, to a circular material economy then ecodesigners in government and business must work together to design and build radical new systems.

Steve Evans is professor of life cycle engineering at Cranfield University. This article first appeared in Inside Track, the magazine of Green Alliance.

Useful links
Cranfield university

See also

• Visionaries: William McDonough - waste guru
• Sustainable cities: rethinking sustainable building

Previous Articles...