Machine Prognostics - Monitoring machine health without humans in the loop

Machine Prognostics’ vision is to empower maintenance engineers all over the world by providing them with reliable actionable information about the health status of their machines.

The company has, together with GPMS Inc., developed a Health and Usage Monitoring System (HUMS) for the shipping industry, which is called Foresight. This innovation is a practical device for monitoring the health and usage of a machine. The Foresight’s sensors have a unique edge design allowing processing of data on-the-fly.

Surfing the IOT wave

“Predictive maintenance is one of the major industry trends, and we are currently surfing the IOT wave,” Thomas Meyer, CEO of Machine Prognostics, explains when asked what makes their solution possible. It was Thomas himself who came up with the idea of using aerospace technology and apply it to the maritime industry. The company has been part of the development of a certified machine monitoring system able to monitor critical and complex assets such as a helicopter’s main rotor gearbox. Machine Prognostics is now capitalising on its background knowledge from the aerospace sector and applying it to the shipping, oil & gas, and subsea industry. The company has used the technology developed by its part-owner, GPMS Inc., for helicopters as a backbone. With a great deal of re-engineering and adaption work to make it fit to the maritime industry, Thomas believes it gives their company an advantage as GPMS is already a recognised a certified company within HUMS and condition monitoring systems for aviation.

The Foresight for the maritime industry is still at a demonstration level in the fields, but Thomas believes that with the right industrial partners the company will reach SME phase 2 and get market traction. As for now, Machine Prognostics is still a small company with limited access to the market due to its size. However, the team is looking for potential reference customers and partners, and believe they are close to getting the attention they need from potential customers as there are no similar solution available in the maritime sector.

Nothing similar on the market yet

Today the majority of the maintenance procedures onboard ships are organised around a time-based maintenance strategy and the process is orchestrated around pre-defined calendar intervals. Thomas believes that this is the case for at least 90% of the maintenance procedures in the maritime industry. For the remaining cases, a so-called condition-based maintenance strategy is used, however, implemented through human based diagnostics – a very different approach than the one Machine Prognostics has come up with which requires no human diagnostics. Machinery health inspections and maintenance tasks are performed at regular intervals regardless of the health condition of the machinery. “The key advantage with a time-based maintenance strategy is the ease of implementation as a human crew easily can construct a master calendar based on maintenance recommendations from equipment manufacturers and follow it”. However, inspecting machinery or replacing parts regardless of their health condition engenders a waste of time and materials. Thus, time-based maintenance generates exceptional costs. One is highly dependent on the maintenance personnel’s diagnostics capabilities, experience and training. In addition, when humans are used for machinery inspection, a report is logged for archive purposes. Maintenance logbooks are text-based entry systems, which means the logged information regarding the health status of the inspected machinery is described by the means of adjectives. The drawbacks with such an approach are that adjectives are by definition qualitative and cannot be used as a reliable qualitative source of information, and a written log is not a natural digital source of information. Furthermore, human-based diagnostics and judgement vary widely among experts.

Thomas also argues that human machinery inspection is neither a permanent solution nor a cost-effective one. “As long as humans are required to diagnose and prognose machinery health, the maritime industry will not be able to truly capitalise on the opportunities linked to the digitalisation of maintenance data, as the information provided will not be repeatable and fast enough to create truly automated systems enabling subsequent business opportunities,” he continues. The fact that most maintenance procedures still are time-based, creates a great business opportunity for companies like Machine Prognostics, which can provide a more time-efficient and cost reducing solution.

Automated and simple to understand

The Foresight sensors have, as mentioned, a unique design, with the signal condition and data conversation occurring within the sensor package itself, which is a Faraday cage. This ensures that there is no external electromagnetic interference when taking measurements. Additionally, the Foresight sensors have an automated built-in test. If a sensor is damaged and not sending correct data, this data will automatically be discarded. The component diagnostics are performed at the sensor level using condition indicator (CI) algorithms. The information from CIs are fused automatically into health indicators by using advanced mathematical techniques. The health indicators provide a quantified machinery health status which is easy to share and understand. A health index of a machine between 0 and 0.8 indicates that the machinery is healthy and there is no action needed at the moment; an index between 0.8 and 1.0 lets you know that you should plan maintenance actions; when the index is higher than 1.0 maintenance is needed. Maintenance alerts and notifications are automatically set by the system. Furthermore, the remaining useful life (RUL) prognostics models will give a warning at approximately 250 hours’ lifetime and have confidence intervals associated with them. The operator will get a text message or an email when a machinery component is out of tolerance.

“Foresight reduces costly unplanned downtime and wrench time as faulty parts are identified before even opening the asset,” Thomas explains. It also reduces costs as the system is completely automated, have no humans in the loop, monitors the equipment during operation, and therefore there is no inspection requirements that causes the operation to stop to perform the inspection of the machinery. “Our technology can provide a system that avoids introduction of faults by dismantling fully functional equipment and reduce the chance of collateral damage to complex machinery,” Thomas says. The team behind the technology consists of 5 employees, whom are not working full-time, and two consultants. “We have the complete knowledge and can integrate it into various IT systems,” Thomas concludes. This makes Foresight especially well matched for the maritime industry.

Contact Thomas Meyer for more information on Machine Prognostics.

By Martine Farstad