Determination of manufacturing unit root-cause analysis based on conditional monitoring parameters using in-memory paradigm and data-hub rule based optimization platform

Abstract

Different manufacturing plants have their disparate process of conditional monitoring for their processing units with diverse set of sensors which amounts to petabytes of data. This leads to conglomeration of problems limited not only to data management but also lack in ability to present the overall point of view of the critical observations at a real time scenario. Some of these observations impact the business revenue/cost process and due to lack of aggregation efforts, the overview is not available to the decision makers. With the intention of highlighting the critical performance factors and applied techniques we present the overall view point of a platform which will address such issues and assist decision makers with certain data points to make choices leading to profitability and sustainability of their business outlook. The platform will serve as a single point of aggregation for diversified but correlated data points and various custom data logic applied as per the business rules providing a correlation between data. This correlation will help reach an outcome where the user can trace the source of data point of the concerned manufacturing units where technical parameters can be optimized. This data flow and processing will result in root cause identification using the data hub platform and real time analytics can be made available using in-memory column store database approach.