It is now official: ITER has financial difficulties, especially the european consortium is in the center of the crisis. The main problem is the cost overrun: the budget has more than doubled, although the assembly itself has not started yet.
The offiicial press release is here.
“In bringing ITER this far, we have together had to face many daunting challenges. With full mutual understanding, good will and family spirit, we have no doubt that we shall be able to find the best way forward to serve our common objective to secure the complete success of this remarkable project “
In other words, we are in big trouble, we would like to keep the project alive but we do not know how.
The causes of this financial debacle are rooted in both the engineering and management levels.
This is nothing new to say that ITER is a huge and complex project, a “behemoth”, some would lambast, developed only to maintain the jobs of thousands of physicists.
But other huge accomplishments have already been achieved: Apollo, LHC,.. of course not unfettered, but the goals were reached.
Projects like that are made of three ingredients:
- the system itself, which is the interaction of all sub-components . This is the most obvious issue that has to be addressed: how to organize the different components, to determine the environment in which they will interact, to plan the life-cycle and the operations. I recommend the lecture of this NASA’s handbook and the Wikipedia page can also be useful. Curiously, this part of the project, which is its essence, is still very often misjudged: the structure adapted to the development of a system is adopted in the latest stages of the design (more precisely when the design becomes clear and so the associated cost overrun). In addition, the job of system designer is a very specific activity which requires particular competences which are different from those of a product designer: there is a lot of focus on abstract entities like, interfaces, margins, environment… This misconception of the system design lead in the past to great failures like the Europa project (and here too): each country developed its stage of the rocket without taking into account (yes, I exaggerate) the interfaces with the other stages. There is also a methodology associated with the development of a system: documentation, quality assurance, processes. I must say that I have no clue about the state of these tools on ITER.
- the technology: the design of each part of the project must be mastered and understood in the finest details to know the margins, to know its reactions in a given environment. If the behaviour, the reliability, the safety of a sub-component is not clearly apprehended, it would be even more difficult to predict its operation when interfaced with other sub-components. That’s why, a rule-of-thumb in astronautics recommends to use a technology which is at least 10 years old. You won’t find the latest INTEL-chip in the brain of a spaceship. To make things clear, NASA developed the notion TRL: Technology Readiness Level.
- the science behind the technology: for a space mission, it includes aerodynamics, trajectory, material science, … The science serves as support as well as for the development of the technology as for the deisgn of the system.
After having exposed these three components of an engineering project, we can better understand how the costs impact the project: the main part of a project’s budget should be the cost of the system, you cannot avoid it. For a big project, the system itself is a new design: either because it is bigger than the other same types of systems and you have to take into account the change of scale (it is a very important factor for ITER: scaling from JET to ITER leads to a completely different engineering method). Or, because the system is simply new. So the design of the system generally represents the lion’s share of the costs. To limit the spending it is thus interesting to use an existing technology (but it is also interesting for the safety) and to support your project on known physics. As a result, you can focus your research effort on the system’s issues.
This is not the case for ITER.
The system is a problem (because of the size and complexity), the technology is a problem (blanket, diagnostics, radiation) and the physics is a problem: tremendous progresses have been achieved in the last 50 years in the understanding of plasma behaviour in tokamaks; yet some major issues, like the origin of the H-mode, the source of instabilities like ELMs, which are really important for the stability and efficiency of the machine, are still under investigation. As a result, the research has to share its money on ALL parts of the projects.
And the second level of a project: the management.
This part always rises criticism from the part of the engineers: it adds too much friction in the machine: too many papers, meeting, powerpoint presentation. Management is at the service of the development, not the other way round. The management of ITER is pinpointed by a large part of the fusion community as bad, not only because of its structure but also because of its essence: too many countries, too many levels of decision, it is a jumble. These points are true. In addition ITER, and its subsidiaries, especially in Europe, F4E and EFDA, decided to hire most of the top-level physicists to manage the project. Double consequences, now: first we lack senior scientists in the day-to-day life to support the research, second we have at the head of ITER project bunch of persons which are great scientists but, for most of them, no manager at all.
The project can also stumble on the lack of engineers: Tokamaks have been physics experiments, developed by physicists for physicists, with the help of technicians. It was possible because of the limited size of these machines and the limitied amount of safety required. The change in scale induced by ITER transforms it in an engineering project. To do engineering, engineers are probably the best choice. The problem is that there is almost no engineer with a background in fusion technology. The European Union decided some years ago to start several programs to train the engineers. The idea is very good but it risks to be impeded by two facts: firstly, the training program are proposed to both junior and senior engineers to benefit both from the motivation and the experience. But the proposed salaries were so low compared with the industry standards that no senior engineer found it appealing. As a result, most trainees are young people with almost no experience in the industry and they should support the effort of ITER. Secondly, the training programs were started too early: some of them have already reached their end (they are on average 3 years long) although ITER has not yet started. What to do with these freshly trained people. The research institutes cannot afford all of them just to have them wait for the beginning of ITER’s assembly. Consequently they leave the fusion research field for a totally different industry, with the loss of the gained knowledge.
The management, inefficient, heavy, absorbs a large part of the costs and it is frustrating for engineers and physicists to see all this money lost in administrative stuff. And the reaction in this case is to find the responsible for this situation. I am not sure that somebody should endorse the responsibility because the organization does not work; perhaps,there is a more fundamental problem: perhaps, the management theory for such kind of big project does not simply exist. In other words, we do not know how to deal with some many participants from som many different origins so that they can work on so many different topics. We do not know how to handle this extremely high level of complexity. This means that ITER requires a huge effort of research not only in engineering and physics but also in management methods. How interesting a breakthrough in this area would be.
When I contemplate the challenges of ITER, I am filled with two opposite feelings: one of fascination and curiosity for this swashbuckling concept, this is a great hope to make considerable progresses in science and engineering, just profit from it, take as much as you can, learn as much as you can. The other feeling says: it is not time yet to start such a project, we are not ready, it is too ambitious; we should not show arrogance because we still have to learn fundamental things before realizing this high scale fusion reactor. Let us work on smaller tokamaks and improve our knowledge of basic mechanisms. Let us improve our methods of management and research. We still have time for that, we must first learn to save energy, to control our energy needs before trying to release this phenomenal source of power. Yes I am balanced between reason and excitation.