Comparing predictive maintenance models
Launching a predictive maintenance program is rarely a linear journey. While it is simpler to transition from a condition-based strategy, any maintenance team can take their first steps toward establishing a new and competitive strategy for monitoring critical machines. However, it can be difficult to determine the best fit for a facility.
There are many different solutions available, and the right one must fit the unique goals and requirements of a business and the facility. This guide reviews the most common predictive maintenance options. It outlines important factors to consider when evaluating the different options to determine the best solution for your facility.
How to find the best option for a manufacturing facility
Any decision-making team needs to understand the advantages and disadvantages of various predictive maintenance technologies and cost structures to make an informed decision. They need to consider their personnel, budget, and corporate goals to identify the best solution that fits their current needs and future plans.
The three general implementation models for predictive maintenance are:
- Route-based: Manual monitoring with data collection every 30-90 days.
- Intermittent: Automated monitoring with data collection hourly or daily.
- Real-time: Automated monitoring with continuous data collection every few seconds.
For each option, evaluate:
 |
Cost | Initial and ongoing expenses. |
 |
Measurement | Monitoring methods and frequency. |
 |
Data | Sensor data accessibility and ownership. |
 |
Scalability | Program expansion capability. |
Route-based monitoring
Manual data collection with portable devices. Collect data at 30-90 day intervals.
Technicians periodically take measurements with handheld or temporarily-mounted accelerometers. They typically collect data every 30, 60, or 90 days. A vibration analyst reviews the data and generates a report with recommendations.
Cost : Route-based predictive maintenance requires no permanent installations or integration with IT networks. The costs of an in-house operation can vary from $30,000 to 90,000 for the necessary hardware. Then there’s the labor cost: It takes approximately an hour for one person to gather vibration data from a single machine.
Measurement: While offering control over monitoring schedules, manual collection limits data frequency to 30-90 day intervals. This misses sudden changes that can lead to catastrophic breakdowns between measurements. This method is limited to capturing only vibration data, but does not collect enough for comprehensive analysis.
Data: Complete access to machine health data enables in-house analysis, work order creation, and historical trending. This supports development of internal expertise, regardless of whether analysis is outsourced or not.
Scalability: Low equipment costs facilitate program expansion. But, data limitations restrict advancement to more sophisticated predictive maintenance. This method requires additional internal labor resources to monitor more machines, particularly at different facilities.
Intermittent monitoring
Outsourced analysis using cloud-based monitoring with wireless sensors. Collect data hourly or daily.
A third-party provider installs wireless, battery-powered sensors on critical machines. The sensors capture data hourly or daily and transmit it to a proprietary cloud platform. The provider’s analysts review data and send recommendations via digital reports.
Cost : A recurring cost structure with simple installation and no IT integration allows businesses to anticipate and control expenditures. Service plans start around $50,000 and can reach $100,000 per plant. Consultants justify premium pricing by providing specialized software that make full use of cloud-based data. They plan their pricing with the goal of measuring as many machines as possible.
Measurement: This model emphasizes software presentation over sensor capability, resulting in comprehensive reporting from limited information. Automatic transmission enables data collection intermittently with capabilities beyond vibration monitoring. However, battery limitations prevent continuous monitoring, resulting in low-density data.
Data: Most third-party service providers consider sensor data proprietary and only provide final reporting. Manufacturers receive maintenance recommendations with limited opportunity to build internal knowledge about machines, data trending, and maintenance impact. The manufacturer loses access to historical machine data when terminating a contract with their service provider.
Scalability: The cost per machine makes it challenging to scale the service to additional equipment or facilities. No access to data prevents building internal expertise.
Real-time monitoring
Self-managed monitoring and analysis with vendor support. Continuous data collection every few seconds.
This real-time, self-managed monitoring model combines permanently installed wired sensors for continuous data collection with cloud-based dashboard analytics. Artificial Intelligence (AI) tools analyze data, detect anomalies, and generate work orders. While vendors provide initial support, organizations typically aim to transition to full in-house operation using third-party software.
Cost: Sensor installation costs are higher than for portable devices. But they should last for the life of the production line. Implementation demands planned downtime and IT coordination. A typical proof-of-concept project for three machines is around $10,000. Expect to pay $3,000 per machine for multiple sensors and an annual software license cost under $2,500 for 500 data points. Software service agreements (fixes and updates) are less than $2,000 annually.
Measurement: Wired sensors support continuous monitoring for advanced, AI-driven predictive maintenance. Software also supports temperature, flow, pressure, and other sensors along with vibration. Your vendor helps you design your data collection strategy.
Data: Full data ownership ensures the user owns all collected data and can use it however they see fit. This enables comprehensive machine insights, AI model optimization, and unlimited historical analysis.
Scalability: The real-time monitoring model is less expensive to scale than an intermittent monitoring subscription and much more effective than route-based. The most significant consideration is the cost of new equipment for each machine. Using in-house resources and owning dashboards and data reduces costs and makes expansion easy.
At a glance comparison of predictive maintenance models
Choosing a predictive maintenance service model
Most facilities progress gradually from manual measurements to condition-based strategies to real-time monitoring. But, every path is unique. The three factors to evaluate readiness are:
- Current capabilities
- Facility characteristics
- Long-term vision
Each of these factors provide important information that will guide you to the best predictive maintenance service model for your current and future goals.
Capabilities
Assess these resources:
- Technical expertise level
- Maintenance staffing
- IT resources
- Budget allocation
Understanding current staffing and internal resources will help you to select the best model for your current capabilities.
Facility
Evaluate your operation:
- Operational scale
- Equipment criticality
- Current maintenance issues
- Available infrastructure
To determine the scope of the project, evaluate how many machines or locations will need monitoring and what manufacturing efficiency metrics will determine success.
Vision
Consider your goals:
- Growth trajectory
- Knowledge development
- Systems integration
- Data management
Understanding the estimated growth trajectory, scalability of the different models, and plans to develop internal resources will help you determine the best strategy to achieve your business objectives.
How to get started
The best predictive maintenance strategy is simple to implement and maintain. It also prepares a facility for the future. The best way to create a customized preventative maintenence program is to start with a proof-of-concept project. For a free consultation with a Solutions Architect, fill out the form below.
For immediate assistance, contact our service center:
info.us@ifm.com
800-441-8246