Even many IT experts say they are baffled by the complexities of predictive analytics. But Paris-based Pertinence says it has gone to great lengths to ensure that its software is intuitive, enabling non-technical staff to use it after only a small amount of training.
At the touch of a button, employees can discover the consequence of changing a particular part of the manufacturing process, says company president Dave Brock,
thus reducing the large number of faulty products or components that manufacturers typically produce.
The company’s software also helps users to understand the reasons why a process, such as the manufacture of a car windscreen, might have failed, says Hubert de La Raudiere, Pertinence’s vice president of business development. Users can extract the data from any process and generate the underlying rules that govern the data to predict the optimum performance for a process, he says.
Pertinence, which was founded in 1999, has a strong heritage in the manufacturing sector. The company’s core product, ‘Rule Maker’, performs predictive analysis on structured and unstructured data in a data warehouse. It uses data mining techniques developed by Persistence co-founder and CEO Augustin Huret, a veteran of several manufacturing giants including tyre maker Michelin, US industrial conglomerate Honeywell International and GDX Automotive, the rubber manufacturing subsidiary of GenCorp.
Although de La Raudiere claims that Persistence’s software costs far less than similar packages from rivals, including SAS Institute and SPSS, the company has made a slow start, attracting only around a dozen customers so far, two of which are Huret’s former employers.
Observers say that Pertinence also needs to add to its venture capital funding of EU6 million in order to beef-up its sales network and expand its international operations. The company has around 30 employees in France and one representative in the US, but it says it has plans in future to open offices in the US, UK, Germany and Italy.