Managing Automation - October 2008 - (Page 37)

or even a burst of organic growth, for example, organizations are invariably left with disparate data silos containing information about customers, products, and suppliers. “The basic problem is that you have multiple applications within an organization that have been built for a very specific purpose, like inventory management, order management, or planning, each running off a separate database,” says David Corrigan, master data management product manager at IBM. “There are multiple repositories of certain types of data and multiple business processes for which the data is changed and updated, and they’re not all in synch with one another.” Companies also often struggle with regulatory compliance and the need to sort out data about tangled webs of customers and suppliers, says David Feshbach, global master data management competency leader at HP Services. “Not only are there problems with knowing the quality of suppliers, but, in some cases, suppliers are also customers,” he says. “There are synchronization and data quality problems in certain countries where systems don’t provide quality reference data needed to establish one version of the truth.” A common scenario is a global manufacturing company running separate ERP systems in each country in which it operates, even though it may be manufacturing and marketing one product or set of products throughout the world. Take the case of consumer electronics maker Panasonic, which looked to MDM to help manage new product introductions. The company’s product introductions were being held up for weeks at a time, Corrigan says, while its marketing staff worked to gather and organize product data residing in various systems throughout the organization. IBM’s WebSphere Product Center (now known as IBM InfoSphere Master Data Management Server for Product Information Management), which is used to manage the process of creating, enhancing, and distributing product information to customers, dealers, websites, and other applications, provided Panasonic with a central point for product description data. Using the tool, Panasonic cut down the introduction process by two weeks, or 10%. The result was a 2.5% increase in profitability, Corrigan says. “In the consumer electronics space, if you can get a new flat screen TV on the retailers’ shelves a week earlier than the competition around holiday season, you have a real competitive advantage,” Corrigan says. MDM can seem unwieldy because there is no one-size-fits-all product or technology to install, and strategies can vary from company to company. Often different technologies work together to en- able an MDM architecture, including servers, repositories, data hubs, middleware, and integration technologies. The core of the system usually comprises a data repository, deployed alongside an integration or middleware layer, says Pascual Laik, vice president of MDM product strategy at Oracle. Data quality and enrichment tools can be added as well. “MDM essentially is an application that does three basic things,” Laik says. “It consolidates multiple sources of data, cleanses it, and makes it available to share across all the consuming applications throughout the enterprise.” THE IMPORTANCE OF PEOPLE Although there is no turnkey master data management solution, companies should bear in mind some common success factors when embarking on an MDM journey, says Brian Ng, worldwide information management delivery excellence lead at Hewlett-Packard. As is the case with wide-reaching IT projects in general, the role of people trumps the implementation of a particular piece of technology or working with a particular MDM vendor. “Needing support from the business is nothing new, but it’s especially true with MDM because of the sensitivity of the data and because MDM straddles the business and IT realms,” Ng says. Equally important is ensuring that the right people are involved in the governance and stewardship of the master data itself. Finally, a company should follow an established MDM roadmap based on strategic goals as the projKEY PRESSURES DRIVING ect grows, he says. DATA GOVERNANCE INITIATIVES “It’s key to note that a Compliance with numerous regulations lot of data governance and 48% stewardship initiatives are Better risk management about change manageImproved operational management, performance, 44% and disciplines for information growth/complexity ment,” says Aaron Mahi41% Improved accountability mainathan, senior director of platform marketing Strategies aligned to improve results 26% at SAP. “In addition to % of respondents Source: Aberdeen Group, December 2007 tools and technologies, a very strong data governance framework needs to be in place in order to implement any sort of MDM initiative.” In addition to helping to resolve data synchronization problems caused by disparate silos of data in scattered business systems, MDM processes and technologies can help manufacturers keep from feeding bad data into analytics, business intelligence, or manufacturing intelligence applications. “Running analytics applications on invalid data,” Oracle’s Laik says, “negates the value of such systems.” value add 61% 37 October 2008

Table of Contents Feed for the Digital Edition of Managing Automation - October 2008

Managing Automation - October 2008 Contents Take 1 Systems Integration Market Braces for a Wave of Consolidation Kinaxis Launches Program to Lure i2 Customers Patent May Give Mobility a Needed Shot in the Arm New Group Aims at More Efficient Smart Devices Solar, Life Sciences Will Be the Next Frontier For Robots Notes Cover Story:�2009 Companies to Watch Special Report:�Siemens plus UGS: Is the Merger Working? Transformation:�Driving Energy Efficiency Integration: How Clean is Your Data Industries:�Fed Raises Red Flag on Chemicals Product Scan Advertiser Index Next

Managing Automation - October 2008

For optimal viewing of this digital publication, please enable JavaScript and then refresh the page. If you would like to try to load the digital publication without using Flash Player detection, please click here.