Preventive Maintenance for Fleets and Trucks

Most maintenance teams learn PM on fixed assets. Chillers, pumps, air handling units, switchgear: things bolted to a slab that a technician can walk to. Then the same team inherits thirty forklifts, a dozen pickups, four flatbed trucks and a couple of scissor lifts, and the PM program that worked perfectly on the plant room starts producing work orders nobody can complete. The vehicle is on a site forty kilometres away. The forklift has done 900 hours in six months while its twin has done 180. The statutory inspection is due next week and nobody tracked it because it never lived in the CMMS. Fleet PM is not a variation on building PM. It is a different discipline that happens to share the same software, and treating it as an afterthought is why so many organisations run their vehicles reactively while congratulating themselves on a 95 percent PM compliance figure that quietly excludes the fleet.

The message up front: fleet PM works when three things are in place. The trigger is a meter (engine hours or odometer) with a calendar fallback, not a calendar alone. The operator does a documented pre-use inspection every shift and that record reaches the CMMS. And the schedule is built around spare-unit availability so that taking a machine off the line is planned, not an emergency. Get those three right and the checklists almost write themselves. Get them wrong and no amount of telematics will save you.

1. Why fleet PM is a different discipline from fixed-asset PM

The differences are structural, not cosmetic, and each one changes how you design the program.

• Usage varies enormously between identical assets. Two identical 2.5-tonne forklifts bought on the same day can be 700 engine hours apart within a year, because one works a double-shift loading bay and the other sits in a stores yard. A calendar PM treats them the same and is therefore wrong for both: over-maintaining one and under-maintaining the other.
• The asset moves. A fixed asset waits for the technician. A vehicle has to be recalled, released by operations, and driven to a workshop or met by a mobile team. Travel and release time can exceed the wrench time, which changes how you batch and schedule work.
• There is a daily operator inspection layer. No building PM program expects a user to inspect the chiller every morning. Every serious fleet program expects the operator to inspect the machine before every shift. That layer catches more genuine defects than the scheduled PM does, and it is free.
• Statutory inspection regimes apply. Vehicles and lifting equipment are subject to registration, roadworthiness and thorough-examination requirements that carry legal consequences for non-compliance. Fixed plant has some equivalents, but the fleet burden is heavier and the deadlines are externally imposed.
• Downtime has a direct operational substitute. If a forklift is down, another forklift can often do the work. That makes spare-unit strategy a genuine PM design input in a way it rarely is for fixed plant.
• Parts logistics are harder. Filters, belts, tyres and brake components are consumed continuously and in variety. A mixed fleet of five makes across three model years multiplies the part numbers you have to hold or source quickly.

If you are building or rebuilding the underlying program, the general frame still applies: start from the complete guide to preventive maintenance and the schedule-building method, then apply the fleet-specific adjustments below.

2. Meter-based triggers: the single most important design decision

Fixed-asset PM is overwhelmingly calendar-driven because calendar works well enough when duty is roughly constant. Fleet PM is meter-driven because duty is not constant. The three meter types that matter:

• Engine hours: the correct primary meter for forklifts, generators, compressors, telehandlers, aerial work platforms and any machine whose work is not measured in distance. An hour meter reflects actual engine and hydraulic wear far better than a date does.
• Odometer / distance: the correct primary meter for road vehicles. Light commercial vehicles and heavy trucks accumulate wear on the drivetrain, brakes and tyres in proportion to distance, and manufacturer service intervals are published in distance bands for exactly that reason.
• Fuel consumed: a less common but genuinely useful proxy on heavy equipment, because litres burned tracks engine work more closely than hours do on machines with a lot of idle time. If your telematics reports fuel, it is worth comparing against hours to see whether your hour-based intervals are being distorted by idling.

The pattern that works in every CMMS I have implemented is meter OR calendar, whichever comes first. Every fleet PM gets both a meter interval and a maximum elapsed-time interval. The 250-hour forklift service also carries a three-month ceiling, so the machine sitting in the stores yard still gets fluids checked, tyres inspected and the battery serviced before the seals dry out and the brake fluid absorbs moisture. Deterioration does not stop when the machine stops.

The weak point of meter-based triggering is meter capture. A PM that fires on engine hours is only as good as the hour reading in the system, and manual readings are late, transcribed wrongly, or skipped. If readings arrive monthly by spreadsheet, your 250-hour service on a high-utilisation machine will routinely fire at 300 hours or more. That is the real argument for telematics, and it is a stronger argument than any analytics feature: automated meter capture makes meter-based PM honest.

3. The operator pre-use inspection: your cheapest and best control

On a well-run fleet, the daily operator inspection catches more developing defects than the scheduled PM. It is performed by the person who knows the machine's normal behaviour, it happens every shift instead of every few months, and it costs three to five minutes. It is also, for powered industrial trucks, a documented expectation in most jurisdictions' operator training requirements rather than a nice-to-have.

For the inspection layer to actually work, four conditions have to hold:

• It must be short enough to be done honestly. A forty-item checklist gets pencil-whipped. Fifteen to twenty items covering safety-critical function is the practical ceiling for a pre-shift check.
• A defect must have somewhere to go. If reporting a defect means the operator loses the machine and gets blamed for the delay, defects stop being reported. The route from a failed check item to a corrective work order must be fast and blame-free. This is where a clean work order type structure earns its keep: an operator-reported defect is a distinct type with its own triage path.
• The record must reach the system. Paper checklists in a clipboard on the machine are better than nothing and worse than useless for trending. Mobile capture, even a simple form, turns the inspection layer into data.
• Supervisors must audit it. Spot-check a sample of completed checks against the actual machine condition. If an inspection sheet says tyres good and the tyre is visibly cut, that is a management problem to fix immediately, because once operators learn the sheet is not read, the whole layer collapses.

Below is the forklift pre-use inspection I would deploy as a baseline. It splits into an engine-off walkaround and an engine-on functional check, which is the sequence that actually makes sense to an operator standing at the machine.

Forklift daily operator pre-use checklist

Two design notes on that table. First, the hour meter reading is captured on the check, every shift, by the operator. That single field solves most of your meter-capture problem before you spend anything on telematics. Second, the rejection criteria are written into the checklist rather than left to judgement, because "tyres OK?" produces a tick and "cuts, chunking, wear to marker" produces an inspection. For more on writing check items that actually get performed, see the PM checklist templates guide.

4. Forklift periodic PM schedule (hour-based)

Beneath the daily operator check sits the technician PM schedule. Forklift manufacturers publish their own intervals and those always take precedence for warranty and for the specific model, but the industry convention clusters around a 250 / 500 / 1000 / 2000 hour structure, with each larger service including everything in the smaller ones. The schedule below is a practitioner's baseline for an internal-combustion counterbalance truck, with the electric variations noted.

5. Statutory inspection: the part that varies by jurisdiction

This is the section where I have to be careful, because the requirements genuinely differ by country and sometimes by emirate, state or province, and getting it wrong has legal consequences rather than just reliability consequences.

The general shape of the obligation is consistent even where the detail is not. Most jurisdictions impose some combination of:

• Road vehicle registration and periodic roadworthiness testing, typically annual for commercial vehicles, often more frequent for heavy goods vehicles and passenger-carrying vehicles.
• Thorough examination of lifting equipment by a competent person at defined intervals. In the UK this is the LOLER regime, commonly six-monthly for equipment lifting people and twelve-monthly otherwise. Many Gulf and Asian jurisdictions apply broadly comparable third-party inspection and certification regimes, often via approved inspection bodies.
• Operator competence and certification records for powered industrial trucks, cranes and aerial work platforms, with periodic refresher requirements.
• Pressure system and gas cylinder inspection for LPG-fuelled units and any accumulator or air receiver on the machine.
• Emissions and tachograph or driver-hours compliance for road fleets, with recording obligations that sit adjacent to maintenance but are frequently managed by the same team.

For authoritative detail, go to the source rather than to a blog. The UK Health and Safety Executive publishes the definitive guidance on lifting equipment examination and on workplace transport at hse.gov.uk , and the US Occupational Safety and Health Administration publishes the powered industrial truck standard including the requirement to examine trucks before each shift at osha.gov . For the international standards on industrial truck operation and maintenance, the relevant series sits with ISO .

Practically, I recommend flagging statutory work orders distinctly, blocking deferral without an authorised override, and driving them off a certificate expiry date rather than a meter, because the regulator cares about the date on the certificate and nothing else. Keep the certificate document attached to the asset record so that an audit is a search rather than a hunt through a filing cabinet.

6. Light commercial vehicle PM schedule

Pickups, panel vans and light trucks are the highest-count, lowest-attention part of most mixed fleets. They are usually maintained by whoever the driver happens to trust, service records live in glove boxes, and the first time anyone looks properly is when a resale valuation or an accident investigation forces it. A simple, enforced schedule closes most of that gap.

The distance bands above are a planning template. Manufacturer service schedules, fuel type, and severe-duty conditions (dusty sites, heavy towing, extreme heat, constant short-run stop-start work) all shorten them. In Gulf conditions I would routinely halve oil-change distance intervals on vehicles doing short urban runs in summer, because oil temperature and contamination behave nothing like the temperate-climate assumptions behind the published figure.

7. Heavy truck PM schedule by mileage band

Heavy trucks justify a more structured tiered programme because the components are expensive, the failure consequences are larger, and the regulatory attention is heavier. The convention is an A / B / C / D tier structure keyed to distance, with each tier inclusive of the ones below.

Two points about the bands. First, they are wide because they should be: a long-haul tractor on motorway work and a tipper on a quarry site do not share an interval, and the quarry unit needs the shorter end of every band. Second, oil analysis is what lets you move confidently within those bands. A structured oil-sampling programme on heavy engines and gearboxes is the cheapest condition-monitoring investment in the fleet world, and it earns its cost by letting you safely extend drain intervals on clean units while catching the contaminated ones early. If you want to push beyond that into condition-driven intervention, the predictive maintenance and failure prediction guide sets out where that actually pays.

8. Tyres, brakes and fluids: the three that cause most of the trouble

Across mixed fleets, a disproportionate share of both unplanned downtime and safety incidents traces back to three subsystems. They deserve named attention in the programme rather than being buried inside a generic PM.

Tyres. Tyres are the largest consumable cost in a road fleet after fuel, and they are mostly managed by pressure discipline: under-inflation raises casing temperature, destroys the shoulder and ruins the casing for retreading. The controls that work are a calibrated gauge on a fixed weekly cycle including inner duals and the spare, tread depth recorded per wheel position rather than per vehicle so an alignment problem becomes visible, a minimum tread depth set above the legal minimum so you replace on your schedule rather than an inspector's, and a wear-pattern review at every PM (centre wear means over-inflation, shoulder wear means under-inflation, one-sided wear means alignment, cupping means suspension). On forklifts the equivalent is checking cushion tyres against the wear marker and rejecting chunked or debonded tyres immediately, because a forklift tyre failure under load is a stability problem, not a puncture.

Brakes. Brake work is the least deferrable item in fleet maintenance and the most frequently deferred, because the symptom appears gradually. What makes a brake programme work is measurement rather than inspection: lining or pad thickness recorded as a number at every PM, at every wheel, so wear rate becomes visible and the next replacement is predictable. On air-braked vehicles add pushrod stroke measurement, leak-down testing and air dryer moisture management, because water in an air system causes a large share of valve and chamber failures. On hydraulic-braked mobile plant, brake fluid moisture content degrades faster in a humid climate than the calendar interval assumes.

Fluids. Treat every fluid change as a sampling opportunity: pull the sample before draining, analyse a defined subset of units, and use the results to adjust intervals. Over a year that gives wear-metal trending per unit, contamination evidence pointing at a failing seal or filter, and the data to justify extending an interval to a sceptical finance function. The immature alternative is changing everything on the most conservative interval and never looking at what came out. Coolant is the most neglected fluid on heavy diesel engines and the one whose degradation costs most: test freeze point, additive concentration and pH rather than judging by colour.

9. Downtime cost, spare units and parts logistics

The design input that separates a working fleet PM programme from a paper one is spare-unit capacity. If every machine is committed to production every shift, there is no window in which to perform PM, and the programme will lose every scheduling argument with operations. The fix is structural, not procedural.

• Size the fleet with a float. A fleet with no spare units cannot be maintained preventively. The float must cover the units simultaneously off the line for planned PM plus a realistic allowance for corrective work and statutory inspection. Present it to operations as an availability requirement, because that is what it is.
• Quantify downtime cost per unit class honestly. A forklift down in a stores yard with three others idle costs almost nothing. One down in a single-machine loading bay that holds up dispatch costs real money per hour. Those two should not share a PM priority or a spares policy, and criticality classification is how you make that distinction defensible rather than personal.
• Stock to criticality, not to price. Filters, belts, brake components, hoses and the consumables for your highest-criticality units belong on the shelf; slow-moving expensive assemblies for low-criticality units belong at the supplier. Mixed-make fleets multiply part numbers, which is a real and often unstated cost of buying on lowest tender each cycle.
• Kit the PM in advance. Pre-kitted PM packs per unit class, issued against the work order, are one of the highest-return process changes available in a fleet workshop.
• Batch by location and window. Travel and release time often exceeds wrench time on dispersed fleets. Group PMs by site and by the window when units are genuinely available, then negotiate that window once with operations rather than per work order.

On measurement, the fleet metrics worth reporting are availability by unit class, mean time between failures, PM compliance split between statutory and discretionary, planned versus unplanned work ratio, and cost per kilometre or cost per operating hour. That last one is the number that makes replace-or-refurbish decisions honest, and it is the number most fleets cannot produce because their costs are not attached to the unit. The wider frame for choosing and reporting these sits in the KPI framework guide.

10. Telematics: what it genuinely changes, and what it does not

Telematics is the most useful technology in fleet maintenance and the most misrepresented. The genuine changes it makes to a PM programme are narrower than the marketing suggests, but they are real and they are valuable.

What it genuinely changes:

• Automated meter capture. This is the big one. Engine hours and odometer flowing automatically into the CMMS makes meter-based PM accurate instead of approximate. It removes transcription errors, removes the lag between the meter passing the trigger and anybody noticing, and removes the argument about whether a PM was late. If telematics does nothing else, this alone justifies it on a fleet of any size.
• Fault code streaming. Diagnostic trouble codes arriving in near real time let you triage remotely: decide whether the unit continues to end of shift, returns now, or stops immediately. That decision used to wait for the driver to notice a lamp and call in.
• Idle time and duty-cycle visibility. Knowing that a unit's hour meter is 40 percent idle changes how you interpret its hours, and sometimes tells you the machine is in the wrong place rather than needing maintenance.
• Driver and operator behaviour data. Harsh braking, over-revving, impact events on forklifts. This is maintenance-relevant because behaviour drives wear, and impact detection in particular surfaces damage that would otherwise be reported by nobody.
• Location and utilisation. Which unit is where, and which units are barely used. Utilisation data is often the strongest argument for fleet right-sizing you will ever get.

For how the integration itself is built, and where OT-to-enterprise data flows usually break, see the IoT fleet management integration guide.

What telematics does not change: it does not tell you whether the brake linings are within limit, whether the mast chain has stretched past its wear tolerance, whether a fork heel is worn, or whether a weld has cracked. Those require a person with a measuring tool. It does not fix a bad PM task list, it does not create workshop capacity, and it does not make operations release a unit for service. And it does not substitute for the operator inspection: the human standing at the machine still catches the cut tyre, the weeping hose and the missing fork pin that no sensor reports.

11. Setting fleet up properly in the CMMS

Most fleet PM problems that look like maintenance problems are actually configuration problems. The setup decisions that matter:

• Every unit is an asset with a permanent ID. Registration or plate numbers change; the internal fleet number should not.
• Meters are first-class objects on the asset. Hours, odometer and fuel where relevant, each with a reading history, a rollover rule, and a validation rule that rejects impossible jumps. A mistyped odometer reading that jumps 200,000 km will fire every PM on the unit at once and then suppress them all for years.
• Meter-and-calendar PM triggers with tolerance. As covered in section two. Configure both, always.
• Statutory PMs are a distinct, non-deferrable class driven off certificate expiry, with the certificate attached to the asset.
• Operator inspection defects have their own work order type with a defined triage SLA, so they do not get lost among planned work.
• Costs post to the unit. Labour, parts, external workshop invoices, tyres, fuel where you can get it. Without this you cannot produce cost per operating hour, and without that number your replacement decisions are guesses.
• Nested PM hierarchy. The 1000-hour service should suppress the coincident 250 and 500 hour services rather than generating three overlapping work orders. Every serious CMMS supports this; many implementations never configure it, which is why compliance reports show phantom overdue PMs.

On platform choice, fleet is one area where the general-purpose CMMS and the specialist tool genuinely diverge. IBM Maximo, Infor EAM and Hexagon EAM handle meter-based PM, statutory inspection and cost rollup well and are the right answer when fleet sits inside a broader asset estate. Mid-market tools such as Fiix, Limble, eMaint, MaintainX and UpKeep handle straightforward meter-driven fleet PM competently and are far faster to stand up. Dedicated fleet management systems do fuel cards, driver compliance, tachograph data and licence management natively in ways a CMMS will not. If both a CMMS and a fleet system are in play, decide once and in writing which one owns the meter and which one owns the work order, because the most common failure mode is two systems each holding half the truth.

12. A practical sequence for building the programme

If you are starting from a fleet that is effectively unmanaged, the order of work matters more than the ambition.

• Step 1: build the register. Every unit, with make, model, year, fleet number, registration, current meter reading, location, custodian and criticality. Building it usually uncovers units nobody was maintaining at all.
• Step 2: establish the statutory baseline. Per unit: what is legally required, when the current certificate expires, who the competent body is. Fix expired certificates before anything else. This is the only part with legal exposure.
• Step 3: launch the operator inspection layer. Short checklist, defect route, meter capture, supervisor audit. Cheap, valuable in the first week, and it fixes your meter data problem as a side effect.
• Step 4: apply a template schedule per unit class. Group by class and duty, apply the template, adjust the outliers. Bespoke schedules per machine are how a programme never launches.
• Step 5: negotiate the maintenance window. Agree with operations, once and in writing, how many units of each class can be off the line and when. This determines whether the schedule is achievable.
• Step 6: kit the PMs and stock to criticality.
• Step 7: automate meter capture. Once the manual programme runs, telematics makes it accurate. In that order, so you are automating a process that works rather than one that does not exist.
• Step 8: review quarterly with data. Compliance, availability, unplanned ratio, cost per hour or kilometre by unit. Adjust intervals on evidence. A schedule that never changes after go-live is one nobody is reading.

Steps one to five cost almost nothing beyond effort and are where most of the reliability gain comes from. The technology in step seven is an accelerator on a working programme, never a substitute for one. For the broader strategy question of which assets deserve which approach, the preventive maintenance strategies guide covers the time-versus-meter-versus-condition decision in more depth.

The idea to walk away with

Fleet PM is a meter-driven, operator-fronted, window-constrained discipline. Those three characteristics, not the checklists, are what make it different. The meter drives the trigger because usage varies wildly between identical units. The operator fronts the programme because they see the machine every shift and the technician does not. And the window constrains everything because a unit that operations will not release cannot be maintained, no matter how good the schedule looks in the CMMS.

The checklists in this article are genuinely usable, and I would deploy them as a baseline tomorrow. But a perfect checklist attached to a calendar trigger on a fleet with no spare units and no operator inspection layer will still fail. Fix the structure first, then the content. The forklift PM checklist is the visible output of a programme; the meter discipline, the defect route and the negotiated window are what make it happen.

Final thoughts

The most common fleet maintenance situation I encounter is not a bad programme. It is no programme: a register that lives in three spreadsheets, service records in glove boxes, statutory certificates tracked by whoever remembers, and a maintenance function that measures its PM compliance on the fixed plant while the vehicles run to failure quietly in the background. If that describes your situation, the single highest-return action is not buying software. It is spending a week building an accurate register with current meter readings and certificate expiry dates, because everything else in this article depends on it and none of it works without it.

Then start the operator inspection. It is the cheapest control available, it produces your meter data, it catches defects nothing else catches, and it changes the relationship between the people who use the machines and the people who look after them. Everything after that, the hour-based schedules, the oil analysis, the telematics integration, is refinement on a foundation that already works. And check your own jurisdiction's statutory requirements before you configure a single inspection interval, because that is the one part of this programme where being approximately right is not good enough.

Related reading: Preventive maintenance: the complete guide, PM checklists, templates and examples, How to build a PM schedule, IoT fleet management integration, Asset criticality classification, Work order types in CMMS.