The Genesis IIoT Platform
Multi-Plant Integrated Reliability and Equipment Monitoring
Fully automated cost-optimal decision making for every machine, system and plant across your entire enterprise.
Cost Recovery Engines
These software engines link plant wide monitoring to operational, supply chain, and maintenance decision-making. Anomalies and prognostics enable automated and systematic cost extraction from on-going operations. The results are higher uptime, fewer emergency service events and extended equipment life.
NEXT-GEN PREDICTIVE ANALYTICS, PROGNOSTICS
Remaining Useful Life
EXTEND ASSET LIFE
Most existing predictive technologies are based on adhoc methods that only generate short-term predictions, which are often inaccurate. These typically include simple trend analysis or physics-based first principles approaches. Genesis proprietary prognostics algorithms leverage uncertainty quantification methods to generate long-term statistical predictions of remaining useful life (RUL) with probabilistic risk profiles and confidence intervals, which describe an asset’s health as soon as it is put into operation.
Existing prediction algorithms take a simplistic approach and assume a single failure mode. In reality, capital-intensive equipment are complex and can fail in different ways (failure modes). They are often monitored by different types of sensors and operate under varying operating conditions. Genesis prognostics technology utilizes advanced data fusion algorithms combined data from different sensors enabling that allow for anticipating the likelihood of a given failure mode and predicting the associated RUL.
RUL predictions need to be adaptive, meaning evolving as external conditions change. Genesis’ real-time on-line learning capability utilizes in-situ data from equipment in operation to update and revise fault diagnostics and RUL predictions based on most recent equipment health. Algorithmic computations can be performed in real-time and in big data settings to support the entire enterprise.
Automated Decision Making
The Genesis IIoT Platform automates decision making by integrating prognostics with supply chain cost-optimal decision making. By converting unexpected, or emergency repairs to planned events, large stocks of just-in-case spares inventories can be reduced in size because parts can be ordered on a just-in-time basis.
DECISION MAKING ENGINE
REDUCE SUPPORT COSTS
Most industrial maintenance policies are based on periodic schedules. Enterprises that have utilized condition based maintenance (CBM) only utilize fault detection methods to trigger maintenance. In contrast, Genesis' CBM algorithms integrate long-term RUL predictions with economic models to obtain cost-optimal maintenance schedules for different loads and settings including opportunistic maintenance policies and fleet-management.
REDUCE SPARES INVENTORIES
Logistic and supply chain software typically focus only on raw materials and finished goods where demand functions are well-understood and can be estimated using market data. In contrast, service logistics or spare parts supply chain is still in its infancy due to a lack of understanding of how equipment reliability interacts with spare part demand. Genesis Adaptive Service Logistics utilizes sensor-based predictive analytics to estimate demand of spare parts and develop cost-optimal spare part ordering policies and optimized inventory management to create a more efficient service organization.
IMPACT OF LOADING
Plant-wide operational decisions such as loading profiles, throughput rates, production schedules, etc. have a significant impact on equipment degradation rates, which is typically overlooked in most existing software solutions. Genesis’ algorithms integrate RUL predictions with operational and planning decision support systems to allow operators to evaluate the impact of a particular loading profile on the asset’s RUL.
HIGH DETECTION SENSITIVITY
Fault detection algorithms are often plagued with false alarms, which typically increase exponentially with the number of sensors being monitored. Advanced data fusion and variable selection algorithms are used to minimize and control false alarm rates while harnessing high density sensor observations to improve detection sensitivity.
ENHANCE REPAIR PROCESS
Effective fault diagnostics are essential for engineering and service personnel. The effectiveness of the diagnostic algorithm is a direct function of the accuracy of the governing fault detection algorithm. Genesis utilizes proprietary machine learning methods and powerful feature extraction capabilities to ensure high-fidelity diagnostic capability.
Industrial cybersecurity solutions are often limited to computer network intrusions. Genesis cyber detection algorithms are complement network defense by identifying network intrusions that target industrial controllers and other SCADA devices such as replay attacks, stuxnet type cyberattacks, and cyber-induced device faults.
Data Curation Engine
Equipment, operational and maintenance data from distributed assets are connected and remotely monitored by Genesis. Reference systems (digital twins) are created and compared to historical data. Failures are identified and fed into the cost recovery engines.
DATA CURATION ENGINE
UNIFY SILOED DATA SETS
Enterprises typically operate equipment with siloed data sets separated by supplier, department, operation, and maintenance strategy. To gain visibility across the entire system, these datasets need to be carefully integrated and jointly analyzed to enable comprehensive system-wide operational visibility.
QUALITY IN, QUALITY OUT
Poor data quality adversely impacts the accuracy of any analytics solutions often generating significant false alarms. Data dropouts due to harsh industrial environments and frequent outliers due to varying data acquisition speeds are just a few examples of poor data quality.. Genesis Data Quality algorithms contain knowledge-based data curation libraries and imputation algorithms that allow users to recover missing observations and enhance data quality.
Industrial data often has diminished value in its raw form. Advanced data transformations using statistical machine learning and signal processing methods enable the extraction of high-fidelity degradation and performance based features that can be utilized to improve fault diagnostics and prognostics.
We integrate into your reporting systems and also provide a full range of capabilities from dashboard visualizations, SMS text notifications and an API for custom interface with Tableau and other visualization platforms.