Use Case: Maximizing well throughput and efficiency by avoiding costly nonproductive downtime (NPT)

In the oil & gas industry, a lack of updated information can hinder management from optimizing drilling logistics and minimizing non-productive downtime. With as many as 40,000 data tags at a drilling site, sensors can track constantly-changing flow rates, pressures, and temperatures to adjust drilling parameters based on real-time data.  However, this sensor data is frequently locked in data stores with limited integration and access to other oil well data.

Arundo's Enterprise solution can generate actionable insights from wells and allow for improved decision-making by managers during critical operations.  For example, associated gas from crude production often contains caustic fumes that increase stress on critical components of compressors, a phenomenon that is visible in the sensor data.

Arundo’s Fabric is able to read and detect this condition, providing an alert to engineers before this critical piece of equipment fails, thus preventing expensive and unforeseen downtime.

Actionable insights allow managers to increase supply-chain efficiency, reduce downtime, and improve key financials measures. The consultancy McKinsey & Company estimates that improving production efficiency by a mere 10% can generate up to $260m of incremental profit at a medium-sized brownfield asset.

Maximizing Well Throughput and Efficiency by Avoiding Costly Nonproductive Downtime (NPT)

Use Case: Improved site safety via anomaly detection

From 2003 to 2010, the fatality rate for oil and gas extraction employees was 7 times greater than the rate for all U.S. industries. Small gas leaks may be unnoticeable to on-site staff but pose serious and sometimes fatal health risks. As drilling and production are often performed in hazardous or extreme environments, a lack of timely or reliable communication infrastructure can hinder critical communication between staff and pose significant threats to employee safety.

Arundo's Fabric can collect and analyze streaming data from a variety of sources including sensor data, weather data and other equipment operational data to detect minor changes in temperature, pressure, or other environmental signals and to better inform operational decision-making.  Early detection of leak-related pressure drops, deviations in temperature or pressure caused by equipment or mechanical failures can be detected, reported, and analyzed through a steady network of communication infrastructure not just within one well installation but across the entire Arundo network of connected wells.  

Arundo's Fabric analyzes stored operational data for root causes, generating actionable tasks for future safety protocol. By automating process steps in day-to-day operations and improving production flow analytics, managers can further enhance safety to all stakeholders by eliminating human error.

Use Case: Automated Parsing of P&IDs to assist in Hierarchy Mapping

Large oil rigs can have more than 50,000 sensors. Documentation regarding the location of the components and sensors are typically provided by different equipment hierarchies and a library of piping and instrumentation diagrams (P&ID). Making queries on equipment hierarchy databases is good for identifying to which larger piece of equipment a sensor tag belongs, but these hierarchies do not map onto the causal chain of events of the underlying process.

Arundo has enabled automatic parsing of P&ID PDFs and this is available as an Arundo Extension. This task is non-trivial, as the PDF format is optimized for on-screen viewing and printing, not for organizing the diagram content. Two apparently equal PDF documents may have a completely different internal representation. However, by combining a series of technologies, like vector graphics extraction, image recognition methods, and optical character recognition tools, this Extension extracts the standardized P&ID symbols for sensor tags, equipment types, etc. and identify how these components are interlinked.

Combining the sensor tag information extracted from the P&ID with the equipment hierarchy allows the creation of a coherent digital skeleton of the industrial asset. This representation enables a definition of different metrics between all different sensors and pieces of equipment, which can be used to identify all potentially correlated sensors to any piece of equipment.

Use Case: Critical equipment monitoring for improved vessel uptime

Critical vessel equipment is currently repaired only after failure or at routine time-based schedules. It can cost over $100,000 per day to have a vessel offline, with additional costs due to equipment order delays and lost revenues.

With Arundo's Enterprise solution, fleet owners and operators can shift from time-based maintenance to conditional monitoring, preventing unnecessary downtime saving on meaningful repair costs. Fabric can reduce vessel downtime due to failures, minimize unneeded maintenance and optimize dry-dock planning through predictive condition-based monitoring. Fabric tracks engine performance in real-time and analyzes for anomalies, providing critical information to engineers that can repair the system and assist in locating and rerouting the ship to the nearest port for emergency maintenance.

Critical Equipment Monitoring for Improved Vessel Uptime

Use Case: Real-time route optimization for increased bunker efficiency

Approximately 80% of global trade by volume is carried by sea, yet the maritime industry lags behind other transport industries in terms of analyzing the vast amount of data and information available. Traditional communication between vessels and ports via radios often results in miscommunication between staff and delayed decision-making. Smart ship sensors collecting deep pools of asset and external data can help optimize shipping routes but operators are stuck in an ocean of information with limited actionable insights.  

Arundo's Enterprise solution can help ship operators by enabling insight-enriched communication between vessels and ports, resulting in increased bunker efficiency and significant cost savings.  Rather than relying on radios to communicate critical location and fleet information, Arundo Fabric connected vessels can capture and share real-time information not only about individual vessels but an entire maritime network. Arundo's Extensions can capture and analyze weather forecast data (for early warning signals or unexpected delays) with data from on-site navigation systems - to automatically recommend changes tailored to the capabilities of each ship’s speed and route, significantly reducing fuel costs and improving utilization of voyage days. These insights improve shareholder value through increased operational efficiency, reduced fuel consumption costs, and higher profit margins.

Use Case: Supply chain logistics / potable water routing

Without real-time asset tracking and analysis (both passenger and cargo ships), planning visibility and efficiency of vessel operations remain challenged. In smaller ports where multiple ships may compete for limited space lack of tracking and analysis can result in huge delays, increased wait times and increased operational costs.

Arundo's Enterprise backed by deep learning algorithms provides vessel operators with optimized shipping routes in real-time, reducing transit and delivery times for vessels containing products near-expiration or held at unsuitable temperatures.  Fabric analyzes vessel speed and other KPIs to accurately plan arrival times to ports, and optimize schedule docking windows. For 3PLs, Fabric's deep learning algorithms can help generate real-time pricing quotes based on current fleet utilization, leading to increased revenue opportunities. These data-driven insights result in supply chain cost savings, increased revenue opportunities, and higher customer satisfaction for the entire vessel operations business.


Use Case: Wind turbine conditional monitoring

In the United States alone, there are over 48,000 utility wind turbines representing nearly 50GW of generating capacity. Operational and maintenance costs can account for between 11% and 30% of a wind project’s leveled cost of electricity, a key statistic for operators and developers. Current operations & maintenance performances on wind turbines are either reactive (i.e. repairing after downtime) or time-based (i.e. scheduled maintenance). This often results in expensive downtime after critical pieces of equipment fail (e.g. gearboxes) or unnecessary and costly routine maintenance.

Conditional monitoring using Arundo's Fabric predictive analytics can detect near-failures at an early stage or prior to such events before they occur. Further, Arundo's Fabric can analyze data from sensors placed on gearboxes and other critical turbine components monitor temperatures and track blade performance, notifying staff when repairs are needed based on an asset’s real-time condition. Often times such repairs may be performed remotely through a centralized operating network, eliminating the need to send personnel to off-shore locations. As a result, developers and operators experience lower maintenance costs, a lower leveled cost of energy, and higher project ROI.

Wind Turbine Conditional Monitoring

Use Case: Monitoring utility grids for anomalies to avoid service interruptions

Electric power utilities are managing aging infrastructure, increased maintenance costs, and potential cyber security threats. The economic impact of the 2003 blackout in the US – largely due to aging substation infrastructure – was estimated between $7 billion and $10 billion. Going forward, utilities are faced with the challenge of improving grid stability while minimizing repair and maintenance costs.

With the rapid adoption of smart meters and rapid decline in sensors costs, utilities now have a wealth of information at their disposal. Arundo's Enterprise solution can improve grid stability by using machine learning algorithms to forecast load patterns through weather and other external data to evaluate the need for additional capacity or renewable curtailment.

Fabric can detect even small grid anomalies in real-time, enabling staff to prepare for and prevent unexpected grid downtime in the future. In the event of an anomaly, Fabric can perform immediate root-cause analysis of historical and archived data, provide the necessary information to troubleshoot at a remote network center. In the event of an outage, utility staff is provided immediate and actionable instructions to respond (e.g. dispatching work crews to the exact location). The result is a more stable grid with improved power quality, fewer outages, reduced costs, and increased customer satisfaction.

Use Case: Hydroelectric plant performance optimization

In the US, aging hydroelectric plants generate nearly 70% of all renewable electricity with a majority built between 1930 and 1970. New investments in hydropower are unlikely due to regulatory hurdles and environmental concerns.  However, power plant operators can now look to Arundo's Enterprise solutions to improve operating and financial performance.

Enterprise's integration with sensor data from power plants and out-of-the-box and custom built machine learning algorithms allow operators to track and analyze underperforming assets, perform root cause analysis and suggest measures to improve KPIs. With Fabric, real-time sensor data can not only be collected, tracked, and analyzed at each hydroelectric plant, but the data can be analyzed across the entire network. Q Foresight can determine whether anomalies in specific data are a result of external or internal factors forming a basis on which to make both mechanical and technical upgrade decisions.

Furthermore, Enterprise is compatible with existing utility SCADA systems.  Benefits for operators include greater visibility into operational data, increased plant performance, and higher returns on investment.

Want to know how Arundo can help you uncover data-driven insights from your asset infrastructure? Click on the link below to request a personalized demo.