Data quality is largely a factor of what rules have been defined for creating new records and what systems have been put in place for adherence to those rules. Most companies struggle due to a deficiency in one or both of these, which leads to data elements like item descriptions or classifications become subjective and prone to inconsistencies or errors.
With distributed plants and locations, material data is thereby subject to individual or organizational preferences, creating a fractured dataset that may be inconsistent, incomplete, duplicated or inadequate.
Over the years, we have developed a rich and comprehensive MRO data taxonomy (aka data dictionary), which covers over 5000 commodities, and is suitable for multiple industries. Based on our own internal taxonomy, we also develop custom data taxonomy for clients if needed.
Once the taxonomy is established for a client, that forms the foundation for an initial data cleansing and enrichment program, as well as serves as the basis for an ongoing data governance system .
The taxonomy typically has three main elements:
Here’s an example of how 3 different inventory managers have created 3 different descriptions for the exact same item
|Inventory Manager||Purchase description|
|Peter (Site 1)||SWITCH, PRESSURE 3 BAR 250 VAC: -25 C TO + 70°C SIEMENS XMLB002A2S11|
|Sam (Site 2)||MONITOR, PRESSURE XMLB002A2S11 3 BAR 250 VAC -25 C TO + 70°C SIEMENS|
|Rob (Site 3)||CONTACT, PRESSURE (SIEMENS); PRESSURE: 3 BAR 250 VAC XMLB002A2S11 -25 C TO + 70°C|