Saferock

Healthcare and Government

SafeRock’s deep experience in systems development, Big Data analytics, and data science situates us as an accretive asset for policymaking organizations in the healthcare and government sectors. Our combination of strategy and science allows us to provide best-in-class advisory and innovative IT solutions for our clients. We can help organizations looking to tap into the vast quantity of raw, unorganized data available in data lakes and other repositories.

SafeRock has a track record of success in providing cutting-edge, analytics-based advisory and IT solutions to our clients. We leverage not only our own proprietary IP but also robust mathematical modeling developed by industry SMEs through our connections with leading research institutions.

ANOMALY DETECTION: CASE STUDIES

SafeRock has broad experience in anomaly detection across multiple sectors. We can process huge quantities of unorganized data and create automated management systems to help our clients identify outliers, anomalies, and areas of risk or potential. Our powerful engine helps eliminate costly frictions by spotting and sorting anomalous data in near real-time, allowing management teams to become immediately aware of abnormal deviations in their primary KPIs.

  • Identified Fraud and Waste

    Identified Fraud, Waste, and Abuse through Unified System   Situation
    • Our client wanted us to help them eliminate areas of potential fraud, waste, or abuse, and identify at-risk observations early on in order to increase value and reduce risk.
    Approach
    • We aimed to create a unified, automated system which would keep a record of vendor funds, identify anomalies early on, and place outlier values in separate buckets for “Hall of Fame” and “Hall of Shame.”
    Result
    • Our system successfully detected and sorted anomalies and outliers.
    • We improved our client’s annual capture by 12% and savings by $2.4 million.

  • Enterprise Data Lake

    Delivered Enterprise Data Lake    Situation
    • A client asked us to deliver a unified Enterprise Data Lake to help improve data management, automate the measurement of key performance indicators, and increase savings.
    Approach
    • We de-duped records daily across a 100k x 100k matrix. This resulted in a clean master data management and seamlessly measured incremental ROI and other KPIs.
    Result
    • Our client experienced an annual 12.4% cost decrease and 19.6% sales increase.

  • E&P Anomaly Detection

    Anomaly Detection in the Energy Sector (E&P)   Situation
    • Clients in the energy sector (E&P) asked us to create a unified analytics system to assess oil and gas production performance in wells. This system would aid public and private E&P operators in forecasting oil and gas production while simultaneously identifying outliers.
    Approach
    • We created a single automated system which combined anomaly detection with well productivity decline curves, improved EUR estimates, and generated type curves for all regions. Our system aimed identify quarterly surprises in well production as well as improve forecasting.
    Result

    • Our combined system flagged outliers daily (from a 30-day lag) and provided predictions for over 15,000 wells. We provided highly accurate forecasting ahead for 1, 2, and 6 months: our estimates outperformed those of industry experts 76% of the time.

MODELING AND FORECASTING: CASE STUDIES

SafeRock’s robust mathematical modeling creates accurate, best-in-class forecasts for our clients. We have experience in predictive analytics across several sectors ranging from retail to E&P. Our successful track record is codified in sophisticated, reusable algorithms and models for a variety of applications to provide a “look-around-the-corner” for management teams. 

  • COVID-19 Incidence Forecasting

    Accurately Forecasted COVID-19    Situation
    • Our client engaged us to perform an econometric analysis of COVID-19 causality and create a model that could accurately forecast COVID-19 incidence.
    Approach
    • We analyzed several socioeconomic causes of COVID-19 including income, wealth, education, health, and other demographics and integrated them into a robust, sophisticated predictive model.
    Result
    • We were able to identify variables with significance levels of 95% and 99% in the model.
    • We increased model accuracy to 95%.

  • Supply Chain Optimization

    Optimized Supply Chain for 400+ Metro Markets in North America   Situation
    • A client asked us to help utilize our predictive analytics to help optimize their supply chain and ensure that supply met demand.
    Approach
    • We proactively identified bottlenecks and OOS, and guided management on where to focus their efforts.  
    Result
    • Our client was able to gain market position in 400+ markets against retail giants like Target and Walmart.

  • Metro-level COVID-19 Modeling

    Measured COVID-19 for 2,600+ Metro Areas   Situation
    • SafeRock was engaged to analyze and model COVID-19 case rates in 2,600 different metro areas.
    Approach
    • We statistically ranked all the metros into six broad categories determined by proximity to urban centers. This allowed us to more thoroughly investigate the differences in COVID-19 case rates between different categories.
    Result

    • We created accurate models in 3,100+ counties and clearly identified counties which were highly effective. Our categorical analysis allowed us to determine the relationship between urban centers and surrounding suburbs in terms of COVID-19 incidence and case rates.