While a production line may seem to be operating normally, they may have concealed failures. Perhaps a machine is beginning to show signs of a mechanical failure; maybe a tool is unaccounted for; materials not available at the time of production. Left alone, these potentially expensive failures can cause a production nightmare.
Industrial asset visibility allows organizations to identify the location of equipment and materials, their condition and how they are being used. Sensors and artificial intelligence (AI) can be used to enhance this asset visibility, transforming physical data from measurements and operations into information teams can action.
The goal is not to send more data. The goal is to provide maintenance crews, engineers and operations managers with a clearer view of what is going on in their plants.
What Is Industrial Asset Visibility?
Industrial asset visibility refers to tracking the location, condition, status and utilization of assets in a facility or supply chain.
These assets could be equipment used in production, tools, raw materials, work in process or finished goods, mobile assets and others. Visibility might be the product's location, the health of the piece of equipment, the presence of the item for a particular task or other parameters.
Envision a conveyor that stays in use while its vibration levels escalate. It might be that a simple machine status indicator indicates the conveyor is on, but it doesn't tell you what's happening. A connected monitoring system can give you more information to go on to determine if maintenance is required.
This set of data on position, status and performance can enable organizations to make more accurate decisions regarding operations.
Sensor Data Acquisition Procedures in the Industry
Sensors allow measurements to be taken to control the equipment and conditions in operation. Sensor types include:
- Temperature sensors gauges the heat levels of bearings, motors, electrical equipment, and other components.
- Vibration sensors can detect any alterations that could be related to misalignment, imbalance, wear of bearing and looseness.
- Pressure and flow sensors used to measure conditions in hydraulic, pneumatic and fluid-handling systems.
- RFID and location technologies is used for automation. Using this technology, tools, materials, containers and mobile assets can be identified or tracked.
- The current and power sensors can measure changes in the behavior of the equipment and power consumption.
- Machine vision systems can be used for visual inspections, component recognition, and certain quality-control applications.
They will obviously vary depending on what you are monitoring and what problem a company wants to solve. Simply putting in sensors is merely adding a lot of data without delivering any significant operational benefit.
In addition to sensors, your team should also integrate data from maintenancerecords, inventories, and production plans to gain a comprehensive view of your assets.
Four Ways AI Improves Asset Visibility
1. Identifying Early Signs of Equipment Problems
By monitoring measurements such as vibration, temperature, and power draw, the AI system is able to recognize the patterns that can highlight abnormal equipment operation.
Take, for instance, a motor that slowly begins experiencing higher vibration levels during similar operating conditions. A monitoring system may raise an alarm so a technician can research the issue before the motor unexpectedly fails.
Such alerts cannot ensure the breakdown of equipment is avoided. They are only as good as the sensors, the past data, the conditions of operation, and the effectiveness of the team to react. However, early warnings are still valuable bits of data for maintenance planning.
2. Tracking Assets and Improving Utilization
Searching can eat into your time and affect the throughput of your production schedule. Automated RFID and location-tracking systems can enable teams to see where assets are and log their movement within the facility.
In addition to tracking, the information that is gathered can reveal idle equipment and inefficient flow between areas.
The objective is to be able to find existing resources more quickly and have better control over them rather than adding new resources just because we are not sure what is out there.
3. Supporting Maintenance Priorities
Conventional Maintenance traditionally depend upon predetermined schedules or overdue action following failure of equipment. Condition Monitoring provides a further piece of information by visually indicating the operating characteristics of an asset.
AI-enabled analysis allows maintenance teams to prioritize inspections based on equipment condition, history of operation, and the impact of a failure.
Decisions about maintenance should still be based upon the requirements set out by the manufacturer, safety standards, inspection processes and the experienced opinion of qualified personnel. AI recommendations are decision-support tools, not a replacement for proper engineering supervision.
4. Connecting Asset Data With Production Operations
Asset monitoring is more beneficial if information exists in each of the relevant operational systems.
They could also view equipment status in combination with production plans, inventory status, maintenance history, and quality data. This allows staff to see how an equipment issue may impact other tasks and which course of action is most urgent.
These integrated systems are part of the emerging Physical AI and AIoT domains in which smart software interfaces with physical machines, sensors, and connected objects. [Aperture Venture Studio's write-up on Physical AI and AIoT Engines]https://apertureventurestudio.com/physical-ai-and-aiot-engines-overview-and-applications/) can offer more background on these solutions and their use cases in industry.
Issues to Be Tackled Prior to Adoption
Sensors and AI can enhance transparency but there are a few challenges to be aware of when implementing them.
Data quality:Sensor Measurements taken at incorrect or missing points, or mis-calibrated instruments can compromise analysis. A methodology should be developed to validate and maintain instrumentation quality.
When appropriate, new monitors should be compatible with existing equipment and process controls. Older equipment can often be integrated with some additional instrumentation or communications gateways.
Cybersecurity: Operational technology should always be viewed as a potential entry point. Implementing controls to limit access, segment networks, and monitor all devices on an ongoing basis can limit exposure.
Alert fatigue: Too many alerts can desensitise staff and lead to critical alerts being missed. Alerts should be tiered according to circumstance of use and degree of impact.
Workforce preparedness: Operators and maintenance staff must know what is being monitored, how alerts are reported, and when to schedule additional diagnosis.
A Practical Roadmap for Getting Started
Businesses don't need to monitor all machines right off the bat to see whether asset visibility is worth it. A pilot for a subset of the environment could be a good place to start.
- Select a specific operational challenge. Begin with one such as regularly scheduled downtime, a tool misplaced, a key production line that's poorly monitored.
- Choose the right assets and sensors. Asset importance, current instrumentation, sensor data, and conditions to monitor.
- Track a starting point. Record how long your assets are idle, the cost of maintenance, or how long it takes to find an asset.
- Validate the information. Make sure the data recorded is reliable and that normal variations can be identified from outliers that require consideration.
- Test notifications and processes. Engage the operators and maintenance team to assess if the notices are meaningful, comprehensible, and actionable.
- Evaluate Results First, Then Expand. Improve upon the original baseline, and increase the size of the system only when the pilot proves to be worth doing.
Performance measures such as unplanned downtime, mean time to repair, asset utilization, inventory accuracy, false-alert rate, and maintenance costs are among the common measures that have been found useful. Which ones are most meaningful varies with the base business problem.
Building a More Informed Industrial Operation
The more accurate information you get, the better the visibility. And to achieve this, trustworthy measurements have to be combined with systems that provide people with the ability to interpret information and make decisions accordingly. Sensors can show physical changes, while AI can help prioritize and determine when maintenance and operational decisions should be made.
https://apertureventurestudio .com/
Top comments (0)