Reducing Data Wrangling with Uptake Fusion

We recently announced that Uptake Fusion is supporting Microsoft Azure Data Explorer. What does that change mean for Uptake Fusion users, data wrangling, feature engineering, and getting the most out of industrial data?

In this blog, we explain some of the updates to Uptake Fusion, how it lends visibility into operational risk and productivity, and what users can expect.

Azure Data Explorer was first announced in 2018 as part of the Microsoft Azure platform. It’s a Big Data analytics cloud and data exploration tool that ingests structured (like geolocation), semi-structured (like events), and unstructured (like notes from a work order) data. The goal: to do real-time analysis on large volumes of data.

Uptake Fusion now supports Azure Data Explorer. What that means for the data of asset-intensive companies is that their time series (instantaneous, historical, metadata), events, alarms, asset management, and work order data can be made available in a single place that is ideal for industrial intelligence.

Industrial data feeds into Azure Data Explorer, and by virtue of Azure Data Explorer, quickly into their own or Uptake’s Azure cloud environment.

That’s where the visualization, query, and retrieval capabilities of Azure Data Explorer enrich the target use cases for industrial intelligence. Broadly speaking, Azure Data Explorer enhances data’s range, scale, access, and speed to readiness for industrial data analytics.

1. Range: Uptake Fusion supports the movement, storage, and curation of industrial data: now including events, alarms, asset management, and work order data in addition to time-series.

2. Scale: Uptake Fusion already ingests high volumes of data. Industrial businesses can support both the scale and scope of their data in the cloud with which to support valuable use cases in advanced industrial analytics and asset performance management.

3. Access: Rather than sitting in on-premise systems with limited access for select users, data in Uptake Fusion is available to approved first- and third-party users in Microsoft Azure.

4. Speed to Value: With industrial data organized and curated in one place with Uptake Fusion, industrial companies can quickly develop and apply industrial analytics. That translates in faster time from data acquisition to business value.

Azure Data Explorer has feature engineering built-in, meaning that data scientists and engineers can focus their time on the development of impactful analytics from more available data.

With Uptake Fusion supporting Azure Data Explorer, asset-intensive companies unlock dynamic visibility into operations.

Uptake Fusion extends object models to the cloud for specific use cases like AI/ML, enterprise reporting and auditing, asset monitoring, and digital twins. Integrating events and alarms data through Azure Data Explorer – like unplanned downtime and operational shifts – enable the creation of analytics for enterprise benchmarks and opportunities for operational improvement.

At the same time, Uptake Fusion maintains the integrity of data sources across business units and individual teams with metadata. The movement of metadata to the cloud allows enterprise datasets to reflect the complex organization of assets at an individual facility. Input from engineers and data scientists store critical asset and plant-level expertise in the cloud.

These flexible asset hierarchies in Uptake Fusion allow industrial data to be used, and then consumed again for analytics that meet the needs of individual data consumers.

Engineers and data scientists have typically gained visibility into risk through a manual process of collecting and then organizing industrial data.

In the instance of a bearing failure, engineers and data scientists look to quickly collaborate on understanding the downstream effects on the production process. These information systems can also be time-intensive.

Uptake Fusion ensures that industrial data analytics reflect the operational risk as the dynamic system that it is: between component health and asset and process failures. When the causality of risk is known and can be stored, teams or the enterprise don’t have to recreate the wheel.

Use cases in industrial intelligence can be created with data in the cloud, with less development time and money. For industrial data scientists and reliability engineers, the data integration readies enterprise-scale data for analysis. It limits data wrangling and feature engineering, and moves beyond the siloed, on-premise order of old.

Data scientists and engineers can focus on value-add data analysis: action-oriented insights that uncover improved decisions to be made across the enterprise. And for sustainability initiatives, businesses can maintain the audit trail for compliance and shareholder reporting.

For decision-makers, enterprise-scale data prepared for industrial analytics directs attention toward valuable opportunities for improvement – rather than on the tools to uncover those opportunities. As asset-intensive companies undertake initiatives across their business, the availability of enterprise-scale data for industrial analytics can be adapted to their needs.

Ready to put your data to work?