My Life's Work: From Firefighting with Spreadsheets to Building a Maintenance Plan That Actually Works
May 27, 2025
Maintenance Plan
My name isn’t important. What is important is that for the last twenty-two years, I’ve been a maintenance planner. I’ve worked in food processing, heavy manufacturing, and now in a highly automated distribution center. I’ve seen the evolution from paper work orders and greasy binders to the promise of AI and a truly connected workforce.
And I can tell you one thing for certain: most of the guides you read online about creating a "maintenance plan" are written by people who have never had to face a production manager at 3 AM to explain why a critical conveyor is down.
They’re full of neat theories and tidy checklists. But they miss the grime, the pressure, and the chaos that defines the reality of maintenance. They miss the human element. My world isn't a textbook diagram; it's a complex, living organism of people, machines, and relentless operational demand.
I used to be a glorified firefighter. My "maintenance plan" was a clunky Excel spreadsheet and a whiteboard that was perpetually out of date. My days were a frantic symphony of radio calls, frantic searches for spare parts, and horse-trading with production to get a few precious hours of downtime.
We weren't planning; we were reacting. And we were losing.
This is the story of how my team and I clawed our way out of that reactive spiral. This is the real-world, no-nonsense guide to building a maintenance plan that brings order to the chaos, earns you a seat at the table, and transforms your maintenance department from a cost center into a strategic advantage.
The Turning Point: Admitting the Old Way Was Broken
The breaking point for us came during a brutal summer heatwave. An HVAC unit cooling a critical server room failed. It was a "surprise," but it shouldn't have been. The preventive maintenance (PM) task for that unit was on my spreadsheet, but it had been bumped three times for more "urgent" work. The failure cost us six figures in lost production and IT hardware.
In the review meeting that followed, I laid it all out: our system was built on guesswork, memory, and luck. We had no real data, no true visibility, and no way to prioritize effectively. We needed to stop firefighting and start engineering our reliability. We needed a real maintenance plan.
Here are the exact steps we took. This is the framework that works.
Step 1: Taming the Chaos – Knowing What You Have (The Asset Hierarchy)
You cannot maintain what you don’t know you have. The first, most crucial step is to build a comprehensive and logical asset registry. My old spreadsheet was a flat list of machine names. It was useless.
We needed a hierarchy. Think of it like a family tree for your facility.
- Level 1: Facility/Area (e.g., Packaging Hall, Warehouse Section B)
- Level 2: System/Line (e.g., Bottling Line 2, Palletizer System)
- Level 3: Asset (e.g., Filler Machine, Robotic Arm, Stretch Wrapper)
- Level 4: Component (e.g., Main Drive Motor, Gearbox, HMI Screen)
Why this level of detail? Because failures don't happen to "the line"; they happen to a specific motor on a specific machine. A detailed hierarchy allows you to pinpoint problems, track costs, and analyze failure data with incredible precision.
We walked the floor for weeks, tagging every single asset and building this tree inside our new Computerized Maintenance Management System (CMMS). It was tedious, but it was the most important work we did. For the first time, we had a single source of truth for every maintainable asset in our facility.
Step 2: Defining the Mission – What Does ‘Maintained’ Even Mean?
With our asset list complete, we had to define our strategy for each one. Not all equipment is created equal. The motor on your main product conveyor is infinitely more critical than the one on an exhaust fan in a storage room.
We categorized our maintenance strategies based on asset criticality:
- Reactive Maintenance (Run-to-Fail): This sounds like a dirty word, but it’s a valid strategy for non-critical, low-cost, easily replaceable assets. The plan here is simply to have the spare part on hand and the work instructions ready for when it fails. We applied this to things like light fixtures and small office HVAC units.
- Preventive Maintenance (PM): This is the backbone of any good maintenance plan. It involves performing scheduled tasks (inspections, lubrication, cleaning, parts replacement) at regular intervals (time-based or usage-based) to prevent failures. This was our plan for the majority of our assets—pumps, motors, conveyors, etc.
- Condition-Based Maintenance (CBM): This is a more advanced strategy where you perform maintenance only when condition-monitoring tools tell you it's necessary. This involves techniques like vibration analysis, thermography, and oil analysis. It’s more resource-intensive but helps avoid performing unnecessary PMs.
- Predictive Maintenance (PdM): This was our ultimate goal, the next evolution of CBM. It involves using data and AI to forecast failures before they happen. At this early stage, we knew it was our future, but we had to nail the fundamentals first.
Assigning a strategy to each asset forced us to think critically about risk and resources. It was the beginning of working smarter, not just harder.
Step 3: From a Messy Calendar to a Master Schedule
This is where the planner truly lives and dies: planning and scheduling. My old whiteboard was a mess of competing priorities. We needed a structured, logical system.
Here’s the process we built, all managed within our CMMS:
- Create Detailed Job Plans: For every PM task, we built a template. It wasn't just "Check motor." It was a detailed procedure:
- Safety First: Lockout/Tagout (LOTO) procedures, required PPE. Tools & Parts: A list of every tool and spare part needed for the job. Step-by-Step Instructions: Clear, concise steps, with photos or diagrams where possible. Estimated Time: A realistic estimate of how long the job should take. Skills Required: Is this a one-person job? Does it require a certified electrician?
- Establish a Planning Horizon: We work on a rolling 4-week schedule. This means at any given time, the next four weeks of PMs are planned and ready to go.
- The "Ready Backlog": A work order isn't "ready" to be scheduled until we have confirmed three things: the job plan is clear, all necessary parts are in stock, and the asset will be available. My primary job became managing this backlog, ensuring a steady flow of ready work for the schedulers.
- Weekly Scheduling Meeting: This is a non-negotiable meeting between me, the maintenance supervisor, and the production supervisor. We review the upcoming week’s schedule, negotiate downtime windows, and lock in the plan. This single meeting eliminated 90% of the daily conflicts we used to have.
Step 4: The Game-Changer – How a Modern CMMS Saved My Sanity
I cannot overstate this: you cannot execute a modern maintenance plan on spreadsheets. It’s impossible. Moving to a true, AI-native CMMS wasn't just a change; it was a total transformation.
Initially, it was just about organization. The asset hierarchy lived there. Our job plans and PM schedules were all managed and automated. Work orders were no longer lost sticky notes; they were tracked digital tasks accessible on a technician's tablet. We could finally track labor hours, parts costs, and downtime accurately against specific assets. For the first time, I could generate a report to show the production manager exactly how much our old, unreliable filler machine was costing us, which built the business case for its replacement.
But the real change happened over the last year. The CMMS started to get smarter. It wasn't just a database anymore; it was becoming an analytical partner. This is where we finally stepped into the world of Predictive Maintenance (PdM).
Our CMMS, connected to sensors on our critical equipment, started doing things I used to only dream of:
- It found problems I couldn't see: We have a massive gearbox on our main palletizer that is critical to the entire facility's output. Last quarter, I got an automated alert from the system. It wasn't a work order; it was an "insight." It had correlated a tiny, imperceptible 0.5% increase in the motor's energy consumption with a specific vibration frequency signature from a sensor we'd installed.
- It predicted the future: The alert didn't just say there was a problem. The AI model predicted a 78% probability of catastrophic failure within the next 60 days. It identified the likely root cause as advanced wear on the secondary shaft bearing.
- It prescribed the solution: The system automatically generated a planned work order, attached the schematics for the gearbox, checked our inventory for the bearing and seal kit (and flagged that we needed to order one), and recommended scheduling the work during our next planned annual shutdown.
We opened the gearbox during the shutdown. Sure enough, the bearing was deeply pitted and close to disintegration. The AI had saved us from a multi-day, six-figure unplanned outage.
That single event changed everything. I was no longer just a planner, scheduling work based on the calendar. I was a strategist, using data to intervene and prevent failures before they ever happened. My team wasn't just fixing things; they were executing surgical, data-driven repairs that saved the company a fortune.
My Advice to You: The Modern Maintenance Plan
So, what does a world-class maintenance plan look like today? It’s a living, breathing ecosystem built on a foundation of solid principles and powered by intelligent technology.
- Build Your Foundation: Don’t skip the hard work. Create a detailed asset hierarchy. Define your maintenance strategies. Standardize your job plans. You must bring order to the chaos first.
- Embrace a True CMMS: Get off spreadsheets. A modern, AI-native CMMS is the single most powerful tool at your disposal. It is the central nervous system of your entire maintenance operation.
- Collaborate to Win: Your plan is useless without buy-in. The weekly scheduling meeting with maintenance and production is your most important tool for alignment and success.
- Let Data Be Your Guide: Stop guessing. Use the data from your CMMS to identify bad actors, justify replacements, and optimize your strategies.
- Step into the Future: Don't be afraid of technology like PdM. It's not here to replace you; it's here to supercharge you. It allows you to evolve from a planner who schedules tasks into a leader who eliminates failure.
My job today looks nothing like it did ten, or even five, years ago. I spend less time fighting fires and more time analyzing data and strategizing with my team. We’re still busy, but it’s a different kind of busy. It’s the calm, focused hum of a well-oiled machine. And for a maintenance planner, there is no better feeling in the world.

Guest post
A special guest post by a maintenance planner at one of Factory AI's customers