An aircraft engine component manufacturer was running an intricate operation generating abundant, although underutilized, data. Operating data was sourced from components manufactured across shops, thousands of parts traveling different routes within the facility and detailed timestamps for each serial number. Given their inability to decipher cost reduction opportunities within their data, the customer experienced significant production losses from costly overruns.
Drawing upon Data Science discoveries, the customer was strategically positioned to tactically reduce costs and accurately predict defects.
GE Digital’s Data Science team was engaged to identify cost reduction opportunities for the customer, as well as determine root causes of variability in the manufacturing process. In doing so, the Data Science team leveraged data that had otherwise been confined to silos, including defect data, timestamp data and material cost data, to compute an ideal route for each part, generate heat maps to break down costs and determine root causes of defects.
Use of heat maps allowed the Data Science team to identify specific paint points in the manufacturing life cycle and develop models for leading indicators of possible delinquencies, cost overruns or defects months ahead of predicted occurrence. Drawing upon Data Science discoveries, the customer was strategically positioned to tactically reduce costs and accurately predict defects, thereby adopting a powerfully cost effective and proactive manufacturing approach.