Skip navigation

Data Matching and Cross Referencing

  • Case Study: View the Honeywell data matching case study.

matchInsight is an intelligent database solution that delivers a "lights out" resolution to cross-referencing issues, while providing an intelligent interface to resolve conflicts. matchInsight Filter Interface

  • Provides a simple GUI for users to resolve issues, and establish rules for new and unresolved issues
  • Fast to prototype and implement
  • Seamless and transparent integration with existing systems and applications
  • Written "from the ground up" in Java (portable)
  • An extensible and scalable solution



The Business Problem

Many manufacturing organizations find they are maintaining the same product and parts data in different formats and in multiple locations across the company. In many cases, the products or parts are in fact the same, although they look different: the manufacturer’s name has changed, the manufacturer was acquired by another firm, there are different versions of the same name for the same manufacturer, or the name is simply spelled wrong (an unfortunate side-effect of a common data entry error).

“Cleaning up” this product and parts data can seem overwhelming when manually matching and resolving problem parts against a ‘standard’ data source. Alternatively, matchInsight provides a simple and immediate way to:

  • Identify duplicate parts data
  • Fix or isolate ‘incorrect’ records
  • Address and resolve typographical errors within the part name and related fields

The matchInsight Strength

matchInsight utilizes iWorks' cross referencing engine as the foundation for the data processing activities that must occur in order to "clean up" dirty parts data.

iWorks provides access to all industry standard file types and most database formats. Standard transformation routines are included out-of-the-box, and a powerful scripting language lets the user extend the functionality of the application to deal with unforeseen data problems.

matchInsight Features

Parts normalization

  • Identifies and standardizes part attributes stored within the part name
  • Part attributes are normalized to a standard measurement (e.g. Imperial to Metric)
  • Normalizes manufacturer names (e.g IBM, I.B.M., ibm)

Identification of potential matches utilizing industry algorithms and iWorks’ proprietary cross-referencing engine

  • Generate base part
  • Name matching
  • Compression-based & Nearness-based matching

Probability rating

  • Flexible scoring routines prioritize suggested matches, and rank the probability of a correct match
  • Adjustable score thresholds let the user focus available time and domain knowledge on the most serious problem parts

Flexible sorting and filtering options let the user review and categorize parts data

Resolution recorder

  • Match problem parts, based on suggested matching parts, and record the outcome
  • Update the back-end repository and close-the-loop

matchInsight Match Interface

The matchInsight "Dashboard"

The matchInsight "Dashboard" is a web-enabled portal allowing global users to collaborate in the effort of identifying and cleaning "dirty" parts data.

Significant Business Benefits

  • 5-10 times productivity gain for users over traditional manual methods
  • Significantly reduces the costs to develop, manage and maintain cross-referencing projects
  • Through a standardized approach and interface, matchInsight enables consistent workflow and staff support
  • Provides an open and scalable solution that offers flexibility after installation
  • Combines all the best features of data quality, cross-referencing, and "Custom Programming" software in one environment

For more information, contact: Philip Hume, Identity & Matching Solutions, The Kenna Group. 905-361-4044.