Vibration Analysis

Continuous vs Route-Based Vibration Monitoring

A comparison of permanently installed continuous sensors versus scheduled route-based portable measurements, covering what each detects, where each fits in a monitoring program, and how to decide which assets need which approach.

Key Insight

Route-based monitoring samples a bearing four times a year. If a fault initiates and progresses to failure within six weeks between visits, the route missed it. Continuous monitoring samples every few minutes or continuously and catches what routes miss. The question is not which is better -- it is which assets can afford the gap.

Fault Severity Time Route visit Route visit Route visit Route visit Route visit Fault develops undetected Route-based monitoring leaves detection gaps between visits. A fault developing between routes may reach Stage 3-4 before the next measurement.

How Route-Based Monitoring Works

A route-based vibration program assigns a technician a portable data collector, a defined list of measurement points, and a schedule. The technician walks the route, places the analyzer on each machine at the designated mounting location, records a waveform and spectrum, and moves on. Typical intervals range from monthly for critical assets to quarterly for general plant machinery.

The measurement itself is identical to what a permanently mounted sensor would capture: acceleration or velocity at a specific location on the bearing housing, analyzed in the frequency domain. Route-based programs can detect imbalance, misalignment, bearing defects, looseness, and gear wear with the same precision as continuous systems, provided the fault is present and measurable at the moment the technician arrives.

That last clause is the operational limitation. A route visit creates a single measurement window that might last 30 to 90 seconds per point. Everything that happens between visits is unobserved.

Where route-based programs are established

How Continuous Monitoring Works

Continuous monitoring permanently mounts one or more accelerometers on each monitored asset. The sensor wires to a data acquisition unit or transmits wirelessly to a gateway. The system samples at defined intervals, from every few seconds to true continuous streaming, and uploads data to a platform where automated algorithms process it against alarm thresholds.

When a bearing defect frequency appears in the spectrum, an alarm fires within the next sampling cycle. The maintenance team has minutes to hours to respond, rather than weeks or months. Fault severity can be tracked across every sample, producing a trend line that shows how quickly a defect is progressing and when it is likely to reach a failure threshold.

The measurement physics are the same as route-based. The difference is sample density. Instead of four data points per year, continuous monitoring generates thousands.

Where continuous monitoring is established

The Gap Risk Calculation

The core engineering question for any asset is: how long does it typically take from the first measurable fault signature to functional failure? If that time is consistently longer than the route interval, route-based monitoring has a reasonable probability of catching the fault before failure. If it is shorter, route-based monitoring cannot reliably prevent failure regardless of technician skill.

The calculation requires a safety factor. A fault that initiates one day after a route visit and progresses to failure in five weeks will not be visible on the next monthly route visit. Even if the average progression time is longer than the route interval, the worst-case scenario is fault initiation just after a visit combined with faster-than-average progression. A practical rule: if median fault progression time is not at least twice the route interval, continuous monitoring is the appropriate choice.

Faults most likely to be missed by routes

Rapidly progressing bearing defects. Small, high-speed bearings can deteriorate from early defect frequency appearance to catastrophic failure within days. Monthly routes cannot reliably catch these.

Intermittent load-dependent faults. Some faults, including certain resonance conditions and load-sensitive gear mesh anomalies, only appear under specific operating conditions. If the machine happens to be at partial load or during a transient condition at the moment of the route visit, the fault signature may not be present or measurable.

Faults that progress in stages with rapid final escalation. A bearing may show early-stage defect frequencies for weeks with minimal change, then accelerate to failure rapidly. Route-based monitoring may capture the stable early stage but miss the rapid final progression if it falls between visits.

What Each Approach Detects

Both approaches detect the same fault types. The physics of vibration measurement do not change based on whether the sensor is portable or permanent. Given equal sensor quality and mounting consistency, a route-based measurement and a continuous measurement taken at the same moment on the same machine will produce identical spectra.

The difference is timing and coverage density. Continuous monitoring detects faults earlier in their progression because data is collected before and shortly after fault initiation. Route-based monitoring detects faults that are present and above the noise floor at the moment of measurement.

Fault Type Route-Based Detection Continuous Detection
Rolling element bearing defects (BPFO, BPFI, BSF, FTF) Yes, if fault is present at time of visit Yes, typically at earlier stage due to high sample density
Mechanical imbalance Yes Yes, with continuous trend tracking
Shaft misalignment Yes Yes
Mechanical looseness Yes Yes
Gear wear and mesh defects Yes Yes
Rapidly progressing bearing failure Unreliable if progression faster than route interval Yes, alarms within next sample cycle
Intermittent load-dependent faults Unreliable if fault absent during visit Yes, captures fault when it occurs
Cavitation Yes, if present during measurement Yes

Cost Comparison

The cost structures of the two approaches are almost inverted. Continuous monitoring carries a higher upfront capital cost: sensors, mounting hardware, data acquisition units or wireless gateways, cabling or network infrastructure, and platform licensing. Route-based monitoring requires a portable analyzer and software, which is a fraction of that investment for the first asset monitored.

The comparison reverses over time and at scale. A route-based program covering 200 measurement points at monthly intervals requires significant technician labor hours annually. That labor cost compounds year over year. Continuous monitoring has near-zero variable cost per measurement: once installed, additional samples cost almost nothing in marginal labor.

At sufficient asset count and route frequency, continuous monitoring becomes cheaper on a cost-per-measurement basis. The crossover point depends on technician labor rates, route density, and how frequently each asset needs to be measured. For programs where assets warrant more frequent than quarterly measurement, continuous monitoring often has a lower total cost over a three to five year horizon.

The failure event cost also factors into the comparison. A single unexpected failure on a production-critical motor can produce downtime costs that exceed the entire continuous monitoring installation budget for that machine. Route-based programs that miss a fault have a direct financial consequence that does not appear in a simple cost-per-measurement comparison.

The Human Factor

Route-based monitoring depends on technician consistency in a way that continuous monitoring does not. Measurement repeatability requires the same sensor placement, the same mounting pressure, the same machine operating state, and the same data collection settings at every visit. A transducer placed two centimeters from the standard location can produce a meaningfully different spectrum. A measurement taken during a coast-down rather than steady-state operation is not comparable to the baseline.

Skilled technicians who own a consistent route for years produce high-quality data. Programs with frequent technician turnover, large rotating teams, or measurement points that are difficult to access consistently produce noisier data with more false alarms and missed detections.

Continuous monitoring eliminates measurement variability. The sensor is in the same location at every sample. The mounting conditions do not change. Operating state can be tracked automatically and measurements taken only during defined steady-state windows. The measurement is as consistent as the electronics allow.

This matters for trending. Detecting a 15 percent increase in a bearing defect frequency requires knowing the previous measurement was taken under identical conditions. Continuous monitoring makes that comparison straightforward. Route-based monitoring requires discipline and process control that is difficult to sustain in high-turnover environments.

Comparison Table

Dimension Continuous Monitoring Route-Based Monitoring
Detection gap Minutes to hours between fault initiation and detection Days to months depending on route interval
Cost model High upfront capital; low marginal cost per measurement Low upfront equipment cost; labor cost scales with asset count and frequency
Personnel requirement Alarm response team; analyst to review trends and set thresholds Skilled technician to perform routes; analyst to review data
Fault progression coverage Catches rapidly progressing faults and faults that appear between scheduled visits Limited to faults measurable at time of visit; cannot cover rapid progression between visits
Measurement consistency High; same sensor location and conditions at every sample Variable; depends on technician technique, placement, and machine state
Data volume High; thousands of measurements per asset per year Low; typically 4 to 52 measurements per asset per year
Alarm response Automated alarm on threshold breach; immediate notification Manual review after route completion; no real-time alerting
Installation complexity Higher; sensor mounting, wiring or wireless setup, platform configuration Lower; portable instrument, route definition in software
Scalability Each new asset requires additional hardware; no labor scaling benefit Additional assets add labor time; instrument can be reused
Best fit Critical assets, high-speed equipment, safety-critical machines, high unplanned downtime cost General plant machinery, slow-progressing faults, broad asset coverage on constrained budget

Decision Framework by Asset Class

The decision logic centers on one variable: the ratio between typical fault-to-failure progression time and the planned measurement interval.

High-speed spindle bearings

Typical fault-to-failure time: hours to a few days. A monthly route interval means route-based monitoring provides essentially no protection against spindle bearing failure. These assets require continuous monitoring or very high-frequency manual measurement (weekly at minimum), which is often more expensive in labor than installing a permanent sensor.

Standard industrial motor bearings (3,000 to 15,000 RPM)

Typical fault-to-failure time: two to eight weeks for early-to-mid stage defects, shorter in final stages. Monthly routes have a reasonable probability of detection if the fault is caught in the early-to-mid stage. Quarterly routes create significant risk. The appropriate choice depends on failure consequence: continuous monitoring for motors driving single-point-of-failure positions; monthly routes for motors with installed spares.

Large slow-speed bearings (less than 300 RPM)

Typical fault-to-failure time: three to twelve months. Large paper machine rolls, kiln tires, and similar slow-speed heavy rotating equipment often give long warning periods. Monthly or even quarterly routes have time to detect faults and schedule maintenance. These are well-suited to route-based monitoring. The challenge is that low-frequency bearing defects require specialized low-frequency transducers and longer measurement periods to resolve clearly.

Gearboxes

Fault progression varies widely by gear type, speed, and loading. High-speed gearboxes driving critical processes should be continuously monitored. Moderate-speed gearboxes with known failure histories and accessible spares can be covered by routes at appropriate intervals.

Pumps and fans

These are often good candidates for monthly routes unless they are single-point-of-failure assets in a production process. Cavitation, bearing wear, and imbalance from fouling are all detectable on a monthly schedule in most cases. Continuous monitoring is justified for pumps where failure results in safety consequences, product contamination, or production stoppage measured in hours of lost output.

The Hybrid Approach

Most mature industrial vibration programs combine both methods. Continuous monitoring covers the 20 to 30 percent of assets where failure risk is highest, fault progression is fastest, or failure consequence is most severe. Route-based monitoring covers the remainder.

This is not a compromise: it is the appropriate engineering response to a heterogeneous asset population with different risk profiles. Applying continuous monitoring to every machine in a plant is rarely economically justified. Applying only route-based monitoring leaves critical assets inadequately covered.

The starting point for any hybrid program is a criticality ranking. Assets are scored on failure consequence (production impact, safety risk, environmental risk) and failure likelihood (age, condition history, operating severity). The highest criticality assets go to continuous monitoring. The rest go on routes calibrated to their fault progression characteristics.

Continuous monitoring data can also inform route optimization. If a permanent sensor on a critical motor shows no change over twelve months, the adjacent general machinery on the route may warrant less frequent visits. Data-driven route intervals reduce labor cost without increasing detection gap risk.

See Continuous Monitoring in Practice

Tractian's condition monitoring platform combines permanently mounted sensors with automated fault detection, giving maintenance teams the coverage and alert speed that route-based programs cannot provide on critical assets.

See Mechanical Monitoring

Frequently Asked Questions

What is the main difference between continuous and route-based vibration monitoring?

Continuous monitoring uses permanently mounted sensors that sample every few minutes or stream data constantly. Route-based monitoring uses a technician carrying a portable analyzer who visits each measurement point on a scheduled interval, typically monthly or quarterly. The core difference is the detection gap: continuous monitoring can catch a fault within minutes of it becoming measurable; route-based monitoring can only detect faults that are present and measurable during the scheduled visit.

Which assets should have continuous vibration monitoring?

Assets where the typical fault-to-failure progression time is shorter than the route interval. This includes high-speed spindles, small high-energy motors, assets in safety-critical service, assets with no standby replacement, and assets where unplanned downtime cost is high relative to sensor cost. Assets with slow-progressing faults, long fault-to-failure times, or low failure consequence are generally suitable for route-based monitoring.

Does continuous monitoring detect different faults than route-based?

No. Both approaches use the same vibration measurement principles and detect the same fault types: bearing defects, imbalance, misalignment, looseness, gear wear, and cavitation. The difference is not what faults they can detect, but when they detect them relative to fault progression. Continuous monitoring detects faults earlier in their progression because the detection opportunity is not constrained by a visit schedule.

How do you calculate whether route-based monitoring is adequate for a given asset?

Compare the asset's typical fault-to-failure progression time against the route interval. If the median time from a first detectable bearing defect frequency to functional failure is longer than twice the route interval, route-based monitoring has a reasonable chance of catching faults before failure. If fault progression is faster than the route interval, continuous monitoring is needed. The safety factor of two accounts for faults that can initiate just after a route visit.

Is a hybrid approach common in industrial vibration programs?

Yes. Most mature vibration programs use continuous monitoring on the 20 to 30 percent of assets that are highest criticality or fastest fault progression, and route-based monitoring on the remainder. This hybrid model concentrates the higher upfront cost of permanent sensors where the detection gap risk is highest, while route-based programs cover the broader asset population at lower cost per point.

MechanicalMonitoring.com provides independent educational content on vibration analysis and rotating equipment reliability. Content is for informational purposes only and does not constitute engineering advice. All product decisions should be validated by qualified reliability engineers for your specific application and operating conditions.