Tagup’s cutting-edge analytics platform keeps your operations running smoothly.

  • What is Tagup? keyboard_arrow_right

    Tagup predicts and helps prevent industrial equipment failures. Our web-based platform all varieties of equipment data, builds machine learning models to predict equipment failure, and makes everything available to authorized users via an easy-to-use web application.

    You may use Tagup to track the real-time performance and estimated time to failure for an individual asset or across your fleet. Our platform is also used by service providers and manufacturers to provide additional services to equipment end-users.

  • Why work with Tagup? keyboard_arrow_right

    Tagup’s machine learning expertise was developed at MIT’s Computer Science and Artificial Intelligence Laboratory (CSAIL). The models we deploy to predict equipment failure and “time to event” are best in class when benchmarked against existing methods.

    For certain equipment types, such as various classes of transformers and membrane-based water technologies, we have large existing data sets and models developed. Our experience with these asset classes enables richer analytical models and more accurate failure predictions.

  • How do we work with Tagup? keyboard_arrow_right

    Except for equipment where we have an existing model, we begin with a Proof of Concept project. We call these “Rapid Assessments”.

    The aim is to quickly and cost-effectively determine the value of our platform deployed at scale as part of your operations. The Rapid Assessment requires: equipment metadata (nameplate, internal references), operational data (instrumentation data, often from a historian), and failure/maintenance records (primarily for model validation).

    Our Rapid Assessment team, comprised of data scientist and software engineers, integrate your equipment into the platform and provide a report summarizing model performance. These are compared against performance goals established at the start of the project to determine next steps.

  • Our customers and partners derive value from Tagup in several ways:

    • Reduced downtime

      Business interruption due to equipment breakdown can be expensive, with lost production and idle resources. By predicting failure events, you can take mitigating action, reducing overall unplanned downtime.

    • Reduced capital expenditure.

      By extending equipment life, Tagup enables you to reduce your amortized capital expenditure.

    • Improved maintenance productivity.

      When you manage a large fleet of equipment, it is important to prioritize the right assets for inspection and repair. By identifying equipment that would benefit most from inspection (based on failure probabilities and “reliability return” on inspection), Tagup improves OPEX efficiency.

    • Improved planning and inventory management.

      By knowing which equipment is most likely to fail months or even years in advance, you can optimize purchasing decisions from Day 1.

  • Primary requirements are data. You must have equipment data, including metadata and operating data. You must also have failure and/or maintenance records to validate model performance.

    We are often asked how much data or how many assets are required for the Tagup platform to work. There are no strict requirements, but as a general rule, more equipment is required as the amount of sensor data and failure examples decrease. So, a small fleet of ten well instrumented wind-turbines may be sufficient, but for transformers, we may require several hundred examples (our best performing models are derived from over ten thousand examples).

  • This depends on the application, but we have hit a positive ROI in under two months for certain applications. During the Rapid Assessment, we will set a “validation period” where we test the platform against real-world equipment breakdown. This period is often several months, and can be as long as twelve months for the largest deployments.

  • Rapid Assessment work is done on a time and material basis. Project minimums are approximately $50k for well-scoped, limited scale deployments with readily-accessible data. Our larger projects are several times this amount (with equipment fleets in the thousands). Project with new equipment types also require additional engineering and data science time.

    Production deployments are billed per asset per month. Pricing is dependent on the equipment (integration and data compute costs vary). Discounts are provided for license term and for volume. Please contact sales@tagup.io for more information.

  • We take data security very seriously. In addition to automated vulnerability testing, we use third parties to validate our security architecture. Data is encrypted end-to-end with 256 AES encryption.

    For more information, please email sales@tagup.io for a detailed security brief.

  • No, only data requirements. We can source data from anywhere, including historians, CMMS, ERP, or other data stores. Our team has experience with establishing ETL processes and using a variety of APIs for easy data integration, including those from OSI PI and eWON.

    For large projects, we can provide recommendations on remote monitoring hardware. We have direct experience with a variety of industrial communications protocols, including Modbus, CAN, and OPC UA.

  • Authorized users can access the Tagup web application from any Internet-connected device via cloud.tagup.io

    In the application’s current limited release, you must be invited to the platform by a colleague or by a Tagup administrator.

  • You can reach us in-application at anytime using the Help icon at the bottom-right of your screen. We are also able to provide help if you email support@tagup.io

Have any questions that are not listed here?