Most preventive maintenance programs in heavy industry were not designed. They were inherited, copied from OEM manuals, or assembled over years by different planners working in isolation. The result is a maintenance program that generates scheduled work without measurably improving equipment reliability.
Maintenance Task Optimisation (MTO) is the structured process of reviewing every task in a PM program, determining whether each task addresses a credible failure mode, and retaining only the tasks that are technically justified and operationally effective. The output is a leaner, more targeted maintenance program with full traceability of what changed, what was removed, and why.
Quick Links
What is Maintenance Task Optimisation?
Maintenance Task Optimisation (MTO) is a systematic review of an existing preventive maintenance program at the individual task level. Every discrete maintenance action within every Work Instruction is evaluated against a structured decision logic to determine whether it is value-adding, needs modification, or should be removed.
The distinction between MTO and other maintenance improvement approaches is important. MTO does not start from first principles. It does not ask “what maintenance should we do?” the way an RCM analysis does. Instead, it asks: “of the maintenance we are already doing, which tasks are technically justified, and which are not?”
This makes MTO particularly effective for mature sites with established PM programs that have grown organically over years. The maintenance program exists. The question is whether it is working.
Why PM Programs Accumulate Non-Value-Adding Tasks
PM programs degrade over time for predictable reasons. Understanding these patterns is the first step toward fixing them.
- OEM schedule adoption without review. Equipment manufacturers provide maintenance schedules designed to protect warranty claims, not to optimise asset availability in your operating context. These schedules are often conservative and fail to account for site-specific conditions, duty cycles, or failure history.
- Reactive task addition after failures. When a significant failure occurs, the common response is to add a new PM task to prevent recurrence. Over time, this pattern creates layers of tasks that were never evaluated as part of a coherent strategy. The failure that triggered the addition may have been a one-off event, but the task persists indefinitely.
- Multiple planners, no consolidation. On large sites with multiple maintenance planners, PM strategies are often developed in isolation across functional groups. The result is overlapping tasks on the same equipment, duplicate inspections with different descriptions, and inconsistent intervals for identical failure modes.
- No formal review cycle. Without a structured review process, PM tasks are rarely removed. New tasks are added as conditions change, but obsolete tasks are never cleaned out. The program grows in one direction only.
- Vague task descriptions. A significant proportion of PM tasks describe activities so broadly that a technician cannot determine what constitutes acceptable completion. “Inspect conveyor” tells the maintainer nothing about what to look for, what to measure, or what condition triggers a corrective action.
Typical proportion of PM tasks found to be non-value-adding in a structured MTO review across heavy industry sites
The 5-Point Decision Logic
Every maintenance task in an MTO review is evaluated against five criteria. A task must satisfy all five points to be retained without modification. Failure on any point triggers a disposition: modify, delete, or replace.
| Point | Question | What It Tests |
|---|---|---|
| 1 | Does this task address a credible failure mode? | The task must target a specific, documented failure mechanism. Tasks with no identifiable failure mode are candidates for deletion. |
| 2 | Is the maintenance strategy appropriate for the failure pattern? | Time-based replacement is only effective for age-related (wear-out) failure patterns. Applying scheduled restoration to a random failure mode wastes resources without reducing failure probability. |
| 3 | Is the task interval technically justified? | The interval should be based on failure data, P-F interval analysis, or OEM evidence. Round-number intervals (30, 60, 90, 180, 365 days) without supporting rationale are a strong indicator of arbitrary scheduling. |
| 4 | Is the task description adequate for execution? | The task must describe what to inspect, measure, or replace, what acceptable condition looks like, and what action to take on finding a defect. Vague descriptions produce inconsistent execution and unreliable condition data. |
| 5 | Is this task unique, or does it duplicate another task? | Redundant tasks performing the same function on the same component at similar intervals are consolidated. This is particularly common on sites where multiple planners have built strategies independently. |
Key insight: The 5-point logic is deliberately binary. Each point produces a pass or fail. This removes subjectivity from the review and ensures that every disposition decision is traceable to a specific analytical finding, not a judgement call made in a workshop.
The MTO Process Step by Step
A rigorous MTO follows a defined sequence. Each phase produces outputs that feed the next. Shortcutting early phases weakens the analysis and undermines the traceability that gives MTO its credibility with clients and regulators.
Phase 1: Scope Selection and Baseline
The first step is defining which assets and PM Plans are in scope. On a large mining site, an MTO engagement might focus on a specific processing area, a fleet of mobile equipment, or a critical system identified through an asset criticality assessment.
Once scope is locked, a baseline snapshot captures the current state: total PM Plans, task count, estimated annual labour hours, trade distribution, and frequency spread. This baseline becomes the “before” picture against which all improvements are measured at the end of the exercise.
Phase 2-3: Task Extraction and Data Quality
Work Instruction documents are uploaded and every discrete maintenance task is extracted. Each task is matched to its parent PM Plan and linked to the asset hierarchy.
A data quality validation step surfaces exceptions before analysis begins:
- Red flags (blocking): Tasks with no linked PM Plan, missing asset references, or unresolvable data conflicts
- Amber warnings: Low extraction confidence, safety/LOTO steps that may have been classified as maintenance tasks, high-frequency tasks that warrant early review
Red items must be resolved before the analysis can proceed. This gate prevents garbage-in-garbage-out problems that plague less structured reviews.
Phase 4: Reverse FMECA
This is where MTO differs from a traditional top-down FMECA. Instead of starting with functions and working down to failure modes, MTO works in reverse: it starts with the existing maintenance task and works upward to identify which component and failure mode the task is intended to address.
Every task is assigned a component and failure mode. Where FMECA data already exists for the asset, the assignment draws on that data. Where it does not, the assignment is inferred from the task description and equipment context.
The output is a complete map of task-to-failure-mode linkages. This map is the foundation for the 5-point analysis in the next phase and for the gap analysis that follows.
Phase 5: MTO Analysis and Disposition
Each task is run through the 5-point decision logic. The output is a disposition for every task:
| Disposition | Code | Meaning |
|---|---|---|
| Retain | R | Task passes all 5 points. No change required. |
| Modify | M | Task addresses a valid failure mode but needs adjustment: interval change, description improvement, trade reassignment, or acceptance criteria addition. |
| Delete | D | Task does not address a credible failure mode, duplicates another task, or applies the wrong strategy for the failure pattern. Removed from the program. |
| Create | C | A new task is required to address a failure mode not covered by the existing program. Identified during gap analysis. |
For tasks dispositioned as Modify, the analyst specifies exactly what changes: the revised description, the new interval, the updated trade, and the rationale for each change. This creates the full audit trail that clients need for their CMMS updates and regulatory documentation.
Phase 6-8: Gap Analysis, Analytics, and New Work Instructions
After analysing existing tasks, the program is compared against a component-level maintenance library to identify gaps: failure modes that exist in the equipment but are not addressed by any task in the current program. Missing tasks are flagged and new tasks created to fill the gaps.
The analytics phase produces the value report: labour hour savings, task count reduction, disposition breakdown, and a before-and-after comparison against the baseline captured in Phase 1. New optimised Work Instructions are generated for each PM Plan, incorporating all retained tasks, modifications, and newly created tasks.
Phases in a structured MTO engagement
Decision points per task
Task-level traceability from original to optimised
PM Defect Classification: Understanding Why Tasks Fail
Knowing that a task should be deleted or modified is useful. Knowing why is more useful. PM defect classification tags every non-conforming task with a structured defect code that identifies the root cause of the deficiency.
| Defect Code | Description | Category |
|---|---|---|
| Vague Description | Task description is too broad for consistent execution | Description |
| No Acceptance Criteria | No measurable standard for what constitutes acceptable condition | Description |
| Arbitrary Interval | Task frequency has no basis in failure data or OEM evidence | Interval |
| Over-Maintained | Interval is significantly shorter than failure data supports | Interval |
| Redundant/Duplicate | Task duplicates another task on the same component | Redundancy |
| Strategy Mismatch | Time-based task applied to a random failure pattern | Structural |
| No Failure Mode | Task does not address any credible failure mechanism | Structural |
When aggregated across an entire site PM program, the defect classification produces a Pareto chart that tells a clear story. A typical finding: “71% of PM defects come from three root causes — vague descriptions, arbitrary intervals, and missing acceptance criteria. These are program governance problems, not equipment problems, and they are fixable.”
This reframes the conversation from “we need to cut maintenance costs” to “we need to fix the maintenance program so it actually works.” It also positions follow-on work: Work Instruction authoring standards, PM review cadence, and reliability capability uplift.
MTO vs RCM: When to Use Which
MTO and RCM are complementary, not competing. The choice depends on where the site is starting from.
| MTO | RCM | |
|---|---|---|
| Starting point | Existing PM program | Asset functions and failure modes |
| Approach | Bottom-up: review what exists | Top-down: build from first principles |
| Best for | Mature sites with established, bloated PM programs | New assets, greenfield sites, or complete strategy rebuild |
| Speed | Days to weeks per asset group | Weeks to months per system |
| Output | Optimised Work Instructions with full lineage | Complete maintenance program from scratch |
| Typical result | 30-40% task reduction, 20-35% labour hour saving | Defensible maintenance basis for the asset lifecycle |
For sites with 500+ PM Plans and years of accumulated maintenance history, MTO is usually the right first step. It delivers measurable results quickly and builds the failure mode understanding that informs a subsequent RCM program on the most critical assets.
Technology-Enabled MTO with HsM
Traditional MTO is a manual, spreadsheet-driven exercise. A reliability engineer extracts tasks from Work Instructions by hand, builds a failure mode mapping in Excel, applies the decision logic row by row, and assembles the value report from pivot tables. For a site with 300 PM Plans and 5,000 individual tasks, this process takes three to four weeks of consultant time.
HolisticAM’s HsM Reliability Strategy platform automates the structured elements of MTO while keeping engineering judgement at the centre of every disposition decision. The platform follows the same 8-phase process described above, but replaces manual data handling with:
- Automated task extraction. Work Instruction documents are uploaded in batch. The platform extracts every discrete maintenance task, classifies each by type (inspection, replacement, lubrication, condition monitoring), and links each task to its parent PM Plan.
- Structured data quality validation. Red/Amber/Grey exception handling surfaces data problems before analysis begins, preventing the garbage-in-garbage-out failures that plague spreadsheet-based reviews.
- AI-assisted failure mode assignment. The reverse FMECA step is supported by an AI classification engine that proposes component and failure mode assignments based on task descriptions, asset context, and the HsM component library. The consultant reviews, accepts, or overrides every assignment.
- Automated 5-point analysis. The decision logic is applied consistently across every task, with PM defect codes assigned automatically. This eliminates the subjectivity and fatigue effects that degrade manual reviews after the first few hundred tasks.
- Trade review workflow. A dedicated Workshop Mode presents one task at a time for trade review in a format designed for large-screen facilitation sessions with maintenance and operations personnel.
- Branded value reporting. The analytics phase generates a client-ready report with before-and-after comparisons, disposition breakdowns, PM defect Pareto analysis, and labour hour savings.
The commercial impact: What previously took three to four weeks of manual effort can now be completed in days. This directly increases the number of MTO engagements that can be delivered per year, supporting the throughput needed to scale consulting delivery without proportionally scaling headcount.
What MTO Delivers
A completed MTO engagement produces tangible, measurable outcomes that clients can action immediately.
- Reduced PM labour hours. Removing non-value-adding tasks and consolidating duplicates typically reduces scheduled maintenance labour by 20-35%. On a site spending $2M per year on scheduled maintenance labour, that represents $400K-$700K in annual savings.
- Improved task quality. Retained and modified tasks have clear descriptions, measurable acceptance criteria, and justified intervals. Technicians know what to do, what to measure, and when to raise a corrective work order.
- Full traceability. Every task carries a complete lineage: original description, disposition, reason for change, and final optimised version. This audit trail is essential for sites operating under ISO 55001, regulatory frameworks, or internal governance requirements.
- Gap identification. The library comparison identifies failure modes that exist in the equipment but are not addressed by any maintenance task. These gaps represent unmanaged risk that the current PM program is not detecting.
- Foundation for continuous improvement. The PM defect classification data provides a roadmap for systemic improvements: if 40% of defects are “vague description” issues, the site needs a Work Instruction authoring standard, not more analysis.
Review your PM program effectiveness
If your PM program has grown organically over years, if you are spending more on scheduled maintenance without seeing reliability improvement, or if your technicians are raising concerns about the relevance of their PM tasks, a structured MTO review can quantify the gap and produce an actionable remediation plan.
HolisticAM delivers MTO engagements across mining, manufacturing, oil and gas, and utilities in Australia. Our engineers average more than 20 years of on-site reliability and maintenance experience.