Massive amounts of data are generated from the vast array of sensors on field equipment such as trucks and trains.
Operators are starting to see the potential value of combining powerful analytics tools with massive live datasets from field operations. We call this opportunity Operations Intelligence or OI.
To create value with OI operators we focus on high-value problems. We then apply the right tools against quality datasets, presenting the results in an intuitive and actionable way. In some cases, the results can feed into automated decision logic, but in most cases the results are used to to make faster, smarter decisions.
Analytics tools have been in use for decades for “what happened?” i.e. routine production reporting. Advances in OI are now allowing operators to also drill into “why is it happening?” and “what is going to happen?” This typically involves processing time-series data as it is streamed from the field against pre-defined machine-learning defined rules.
Leading operators are using OI to predict equipment failures and find new ways to improve operations through advanced data pattern matching.
Improving operations and maintenance
Faster, smarter decisions
Earlier prediction of equipment failures
Reducing data collection and preparation time
Business improvement activities
Most operators are generating and capturing huge volumes of data from the field. However, raw data is only part of the story. Data needs to be put in context before it becomes useful for decision-making. Many organisations have issues in contextualising their data.
For a range of reasons the data captured from the field, or the reference (master) data, may be of poor quality. Root causes may include: the devices capturing the data may not be correctly calibrated, or lack of organisational controls in place for maintenance of reference data. Reporting and analytics solutions are only as good as the base data.
Many organisations have invested in a business data warehouse (DW) for analytics of corporate functions. These organisations have attempted, without success, to extend their existing DW into the domain of field operations. The traditional IT-led DW approach is not effective for operations. As a result, many organisations do not have a clear strategy for operational datasets.
Many organisations have hundreds of reports, developed organically over a period of years. These reports have overlapping functionality, conflicting business rules and data sources, and the resultant cost of support is high. Organisations need assistance to consolidate and simplify their reporting assets based on agreed metrics for operations and agreed sources of truth.
Historically, operators have placed little value on operational data, treating it as a by-product of operations not as a valuable corporate asset. OEM vendors have historically held the data, providing the customer with limited visibility. This has become a major issue for operators. Operators need to find ways of getting better visibility and ownership of the data captured by the OEM vendors in order the data can be fully accessed and manipulated to produce useful decision support information.
Organisations typically have a range of analytics tools in use e.g. MS Excel, real-time dashboards, BI analytics tools, data science advanced analytics tools, process historian tools, and more. These tools have overlapping capabilities. Selecting the wrong tool for the job increases up-front and support costs.
ATI’s services in the field of Operations Intelligence are focused on client-side services, assisting operators to make smarter, faster decisions using analytics.
Technology Strategy & Architecture for OI
Metrics Tree, Master Data & ODS Design
Production Reporting Audits
Support & Improvement Model for OI