Europe’s Herschel-Planck science mission, the most expensive and complicated satellite program ever undertaken in Europe, is more than 20 percent over budget and months behind schedule because of instrument development delays, government-imposed work-distribution rules and the inability of European laboratories to produce hardware as promised, according to government and industry officials.
The mission is now expected to cost at least 1.6 billion euros ($2 billion) heading into a crucial integration and testing period during which the possibility of further overruns cannot be excluded, officials said. The planned launch date of February 2007 for the Herschel and Planck satellites, aboard a single Ariane 5 rocket, has been scrapped; the current best-case date is August 2007.
The budget overrun is having a domino effect on other European Space Agency (ESA) science missions, some of which will face funding delays as ESA and its member governments’ space-science laboratories absorb the impact.
ESA science managers, who have been criticized in the past for letting program ambitions get ahead of financial resources, are trying to keep the budget-cascade effect to a minimum as they prepare to make their case to European governments in December for increased space-science funding.
The Herschel-Planck problems, mainly related to development of the five science instruments to be carried on the two spacecraft, have been building for over a year.
Europe’s Science Program Committee (SPC), which sets priorities for ESA’s science budget, was so surprised at Herschel-Planck’s new cost estimates that it created a board of inquiry to assess what went wrong with the program.
The Herschel/Planck Program Investigation Group’s 21-page report, delivered to the SPC May 10-11 at a meeting in Helsinki, Finland, finds no villains in the Herschel-Planck story. Instead, it concludes that Europe’s complicated contractor-agency-research laboratory system is incapable of digesting a project as demanding as Herschel-Planck.
Risto Pellinen, the outgoing president of the SPC, chaired the board of inquiry and has concluded that the SPC itself shares in the blame for not having insisted on credible cost estimates early on, and for not questioning whether the required technologies were mission-ready .
“It turns out we never really had a cost-at-completion figure,” Pellinen said in an interview. “This is the most challenging science mission ESA has ever done, and we all overestimated the ability of the national laboratories to handle this work.”
Pellinen said the SPC, ESA, the national laboratories and the industrial contractors were complicit in refusing to confront the likely costs of the mission head-on, hoping the problems would take care of themselves. They never did.
Under ESA’s current system, the agency acts as program manager for a science mission, financing the satellite’s basic structure, the integration of the instruments, the launch vehicle and mission operations.
It is up to ESA governments’ national agencies, working with their science institutes, to finance and develop the satellite instruments.
This division of labor has been overwhelmed by the complexity of the five Herschel-Planck instruments, each of which is estimated to cost 100 million euros or more.
Until recently, laboratories agreeing to take on satellite instrument development made unwritten, informal commitments to ESA and the SPC affirming their ability to do the proposed work .
In Herschel-Planck’s case, insufficient attention was given to the complexity of the instruments that scientists had ordered for Hershel’s infrared science laboratory and Planck’s cosmic-radiation payload.
Months passed, and several of these laboratories returned to ESA to say they could not do the work on their own. Even after forming into unwieldy consortia — with up to 10 different funding agencies for each instrument — they asked ESA to help out financially. The situation was exacerbated by the decision by NASA and the Danish government not to provide promised satellite components.
To date, ESA has taken about 78 million euros from its science budget to help the national laboratories complete their Herschel-Planck work. It was “the only way to keep the Herschel-Planck project on its feet,” noted the inquiry board’s report, which said the board could not determine whether further financial life buoys will be needed to complete construction of the instruments.
ESA Science Director David Southwood said Herschel-Planck is the latest example of why he recently instituted a policy calling for more-binding commitments by the national laboratories. “There is no single guilty party here; it’s everybody’s fault, including our member states,” Southwood said in an interview. “One lesson here is that we should never have agreed to do these programs in parallel. It was too optimistic.”
Richard Bonneville, chief science delegate at the French space agency, CNES — and SPC chairman at the time of the Herschel-Planck decision — said science institutes always have resisted making formal commitments.
“At the time, I wanted the institutes to sign a piece of paper, nothing complicated, to affirm their commitment to the project, and outlining who will do what,” Bonneville said in an interview. “There were lots of reasons given for why this was viewed as a bad idea.”
ESA’s decision to combine the Herschel and Planck missions into a single program, built and launched at the same time, was intended to save money. The two satellites’ service modules and other components, it was assumed, could be made nearly identical.
Alcatel Space of Paris was named Herschel-Planck prime contractor in June 2001, signing ESA’s biggest-ever space science contract. Alcatel’s work includes substantial contributions from EADS Astrium of Germany and Britain, and Alenia Spazio of Italy.
Some 140 other contractors across Europe were assigned lesser roles as Alcatel, EADS Astrium and Alenia kept to ESA’s tightly scripted geographic-return policy. These rules dictate that work on ESA programs be distributed among member states in close proportion to each individual nation’s contribution to ESA.
These rules have long been known to increase costs. In this case, Alcatel created a complex industrial organization chart with crisscrossed responsibilities that the inquiry board said was an unneeded drag on the program. The board also said it detected among the principal contractors “a sense of mistrust at all levels … together with a feeling that the schedule was never seen as entirely realistic.”
Joel Chenet, head of Earth observation and science programs at Alcatel Space, agreed that the organization was complex but said ESA’s rules on work distribution force prime contractors to do things this way.
“Given the constraints imposed on us by ESA’s policies, I don’t see how we could have done it more simply,” Chenet said in an interview. “The rules are what they are.”
Chenet said he disagreed with the inquiry board’s comments about friction among the main contractors.
The inquiry report gives several examples of contractor difficulties, including this one, which Chenet confirmed: To assure a wide distribution of work in Europe, the Herschel-Planck contractors assigned one company the task of developing substrates for the spacecraft solar arrays. A U.S. company had demonstrated experience in producing these components, but ESA programs, funded by European taxpayers, give preference to European companies.
As a condition of performing the work, the selected company required a separately funded technology-transfer contract, which was agreed to by ESA.
Months later, it was determined that the selected company would not be able to produce the substrates, adding to the Herschel-Planck delays and costs.
The contract has since been cancelled, and the work has been assigned to a U.S. company.