CIOZone.com - Professional Network for CIOs and IT Professionals

The Key Benefits Of P&G's ELN

Like other PLM software packages, Teamcenter came from a vendor of computer-aided design software and started out as a platform for coordinating the exchange of CAD documents. But for the last several years PLM vendors have been trying to design themselves more broadly. They say PLM software can manage the life cycle of any sort of product—that is, the cycle that starts out in early R&D and follows a product through into prototyping, testing, manufacturing, and subsequent refinements.

For P&G's purposes, the key virtue of Teamcenter from this perspective was "the ability to represent multitudes of different objects and modules and the relationships between them," Staresinic says.

Compared with other enterprise software, such as Enterprise Resource Planning systems, PLM also has the virtue of being a tool for new product development, rather than operational efficiency, he argues. "ERP is all about cost reduction, and you can't cost cut your way to top line growth," Staresinic says.

That doesn't mean the ELN system won't save P&G money.

In fact, the system promises to make the process of documenting research work more efficient, while eliminating wasteful duplication, says project manager Caserta. "Our estimate, and we believe this is a conservative number, is that the average large R&D organization repeats about 15% of all the work it does," Caserta says. "Mostly, it repeats work that failed. If you don't know that this experiment has been tried before and failed, you're almost doomed to repeat it." Researchers working on the same problem as their predecessors tend to independently think of the same experiment and repeat it, not knowing that's been tried before.

One reason researchers don't find out about those previous experiments before repeating them is that the information is locked away in paper lab notebooks. As the R&D organization moves to recording experiments in electronic notebooks that will be easier to search, it ought to be able to eliminate about half of that rework, Caserta figures. "And in an organization of more than 8,000 people, a 7% productivity gain translates into an enormous amount of money." In rough numbers, more than $50 million worth, he says.

The average researcher also spends about 10 hours a week methodically recording data in paper notebooks. Through automation, such as the ability to copy the output of an instrument directly into the notebook (or insert a hyperlink to a record in an archive of instrument data), Caserta estimates they can save two hours per week, for another 5% productivity gain.

Overcoming Hurdles

However, as is the case in many IT development projects, there were some bumps along the way—including the fact that many of the project benefits won't be immediate. For example, the value of being able to search for the results of past experiments won't be realized until the system is widely adopted and there are a few years of experimental results recorded in it. And that's going to take a while. The system also has taken longer to get into production than originally planned. And because of budget constraints will take longer than planned to deploy across the whole R&D organization.

Caserta at one time was projecting the technology pilot that had started in January 2007 would be concluded that July. He thought 1,000 users would be live on the system by October 2007, and 3,000 would be on-line by this spring. Instead, the pilot was followed by another pilot project, as P&G worked through the complexities of improving the user interface and attaching electronic signatures to notebook entries. The plan now is for a more gradual roll-out, with the system to reach 1,300 users by July. And for budgetary reasons, it will be held at that level until July 2009 so that the deployment to the next 3000 to 3,500 users will happen in the 2009/2010 fiscal year. (The budgetary issues are nothing out of the ordinary, Caserta says—mostly re-prioritization to allow other, previously funded projects to be completed).

Another hurdle was electronic signatures—and the major reason P&G's implementation took longer than expected, Caserta says.

An electronic notebook is a collection of all documents related to a research project, organized in Teamcenter. As part of the process of checking a notebook into the system, researchers are prompted to add an electronic signature, just as they would routinely sign and date the results recorded in a paper notebook. Reports that traditionally were printed and added to a three-ring binder can be reduced to Portable Document Format (PDF) files, and the notebook also can contain a link to the system of record for that data.

Adding the electronic signatures turned out to be a sticking point.

To ensure that its solution would meet the requirements of high end of its business, P&G decided early on to adopt the SAFE (Signatures and Authentication for Everyone) specification from the SAFE-Biopharma Association. To promote adoption of the technology, members of the group agree not to challenge the validity of signatures that meet the SAFE requirements. In other words, as another member, Johnson & Johnson would be obligated to respect the SAFE signatures on P&G notebooks related to a patent filing, even if it disputes the patent on some other grounds.

The first obstacle was convincing P&G's own legal team that the technology could be trusted with protecting the company's precious intellectual property. After the approval from the lawyers came through, the next step was issuing the actual digital credentials. Beyond setting technical standards, SAFE dictates that every participating organization have a rigorous process for authenticating the identity of employees who will be issued digital certificates—the cryptographic keys used to complete an electronic signature. P&G contracted with Citi Managed Identity Services, a division of Citibank's parent company, Citigroup, to independently verify the identity of each individual being issued a credential in much the same way that the bank might check an individual applying for a loan. Once authenticated, users of the system are issued an electronic security token that plugs into the USB port of a computer. An individual must supply both a password and the cryptographic key contained within the token to complete an electronic signature.

Getting all that in place took time, Caserta says. Even now, the authentication process requires that researchers present themselves for a personal interview before being issued a token, a process that is slowing deployment to thousands of researchers.

"We're using a couple of approaches to getting people started," Caserta says. "One is that we're giving training classes of about an hour and a half, and when people come in for the training they're issued their certificate. That way, when they walk out of the class, they're able to use the system right away." During the pilot project, about two-thirds of the users went through the course, and the rest were issued their tokens separately and either went through online training or simply figured out how to use the system on their own, he says.

As challenging as the electronic signatures have proven, not having them was worse, Caserta says. During the early pilot project tests, before the electronic signatures had the approval of P&G's lawyers, participating researchers had to record their findings electronically, then print and sign a paper version for a traditional bound notebook. "We found that people hated having a hybrid system even more than we thought they would," Caserta says.

Despite the obstacles, Caserta says P&G is poised to deploy electronic lab notebook technology on a larger scale than any other organization he knows of. Most ELN deployments are on the scale of a few hundred users, he says. "The largest I've heard of was at Kodak, which was about 500 people."