Case Study: Role of Analytic Model for Predictive Maintenance of Medical devices

“Knowing that a machine will soon break down and preventing this at the right moment saves companies time and money. To get this knowledge, analytical evaluation of the existing data is decisive.”

Case Study 1:

Large medical devices such as Diagnostic Imaging, CT scanners, and magnetic resonance imaging systems are a major investment for doctors’ practices and hospitals. Unexpected breakdowns don’t just cause huge costs but also jeopardize patients’ medical care. For manufacturers, this means many spare parts must be kept in stock permanently, resulting in high capital tie-ups. If a device breaks down, the technicians have to take numerous spare parts with them to the customer on spec. And if the parts aren’t needed, they have to be thoroughly checked before they can be restocked.

Goals
  • Predict the failure probability of large medical devices
  • Optimize stockholding for replacement parts and operations management
  • Improve customer service and reduce downtimes
Approach
  • Read system logs and record sensor data centrally
  • Recognize patterns based on historical data
  • Calculate failure probability automatically
  • Merge sensor data with information from the data warehouse
Results
  • Central view of a wide range of business information for the first time
  • Fast and targeted evaluation of data for different departments
  • Increase of product and service quality throughout the company

Big data: information from log files

The big data challenge is ever-present and has been discussed again and again over the past years: Data is lying idle in many companies without being efficiently used. All the manufacturer’s large medical devices send log files with system-relevant status information to the respective development department every day. Yet, in the past, this data was only spot-checked and evaluated by experts manually. The information could neither be used across departments nor were targeted analyses with a large data basis possible. In consequence forecasts about a device’s failure probability couldn’t be made. 

For the manufacturer, this implied:
  • Many of the spare parts had to be kept in stock and sent to customers if a device broke down.
  • Restocking the parts that weren’t needed required lengthy checks.
  • There was always the risk of high costs caused by device downtimes because the response times in the service level agreements were tight.
To reduce costs in the long-term while at the same time increasing the service quality, a German company started an initiative to implement predictive maintenance. The first step was to centrally record all the data from the transmitted log files and enrich it with information from the SAP system. Meanwhile, an analytical model was created that could be used to detect recurring patterns in the data.

Automated evaluation: The added value from raw data

Step by step, the project team implemented various software components to transform the usable information and evaluate it. The large devices send log files to the manufacturer via a file subscription system. There, the relevant sensor data and events are read using a Hadoop cluster. The aggregated data is subsequently stored centrally in a Teradata Data Warehouse. For the analysis and further processing of the Hadoop data, the company opted for a business intelligence platform from the software provider SAS. The team generated the analytical models as the basis for predictive maintenance. The performance turned out to be decisive for the acceptance of the predictive maintenance solution among users. 

Case Study 2:
 
Challenge
Failure prediction by human operators requires advanced skills, and a limited number of experts cannot monitor all medical device systems around the world. "Corrective maintenance" for repairs after breakdowns has also become inevitable.

Solution
Hitachi analyzed three years’ worth of sensor data from 100 medical systems and created a mechanism to investigate the cause patterns that lead to device failures. Then machine learning was used to define a normal operational state to achieve successful early detection of abnormalities and changes in status that lead to failures.

Result
Signs of impending failure have been detected several months before a breakdown occurs, and scheduled maintenance before systems break down has been made possible. As a result, downtime (time systems cannot be used) due to breakdowns has been reduced by 16.3%.

Reference: https://social-innovation.hitachi/en/case_studies/mri_predictive_maintenance/

Comments

Post a Comment

Popular posts from this blog

FPGA In Layman''S Terms

Image
FPGA is an Integrated Circuit which can be reconfigured again and again to perform different tasks as desired. For example, an FPGA can be configured to function as a processor, and then it can be reconfigured to function as a Graphics Processing Unit (GPU), or hardware video encoder, or something else. The point is that FPGA can theoretically be programmed to function as any other Digital Integrated Circuit. Think of FPGA as a breadboard for Digital Circuits. All the gates and flip-flops are present on the FPGA, and there are wires going through the complete chip. Circuits are made by connecting these wires to the relevant gates or flip-flops as per intended design. Instead of connecting the gates/flip-flops using a physical wire as in a breadboard, FPGAs has programmable interconnects that can be rewired programmatically. There are dedicated wires/routes for clock signals and only a selected number of FPGA pins are allowed to drive those global clock routing wires. FPGA Vs Microcontr...
Read more

BOARD BRING-UP

Board bring-up is the process of checking and validating a PCB assembly for the first time. The aim is to make sure that the board has been manufactured and assembled correctly and that all of the subsections of the design work as intended. This can be a nervous time for an electronics designer as it is a key proof point that many weeks or months of work has been successful. Note that bring-up is not the same as validating an entire product or system – it is usually limited to validating the hardware. Writing firmware will likely be a part of the process, but only to the extent required to ensure the processor, its peripherals and external circuitry connected to them are working as expected. Board Bring-up process. 1. Visual Inspection: Print out assembly drawing with all the components and test points to have a quick reference during validation. All the components are placed correctly. All the PINs 1 are placed correctly. There are no visible short circuits between PADS. Are pads sold...
Read more

High Speed Digital Design Considerations

Image
The four main consideration of today’s high-speed design:                                                                                                                                                            Ref(1) We need to consider signal integrity , which includes parameters such as Overshoot and Ringing. Need to look at crosstalk which is about EMI coupling between traces on PCB board Timing is about calculating the flight time EMI and radiation  These parameters are related to each other and aff...
Read more