Best practices for analyzing maintenance logs in order to drive down downtime and increase OEE

Introduction

In order to improve your manufacturing operation, learn how to best document your outages and how to setup your organization, in terms of roles and tasks, to make use of your documentation.

Our main takeaways are:

• Your documentation needs to follow some minimum standards
• Technical Service and Production need to document and solve issues together
• Special role needed to track changes and push through actions

We will look at:

• Stop documentation
  • What should be recorded
  • Who should do it
  • When it should be done
• Analysis
  • Process
  • Meetings
  • Roles / Who

Along the way we will mentioned tools that can support and help you with the tasks.

Data: The goal of documentation

To improve your production it is critical to have data to guide your decisions and improvement efforts. Data helps you to find the biggest loss driver to focus on and help you to align stakeholders in the process.

Without appropriate data to support your actions, it’s much harder to get changes through in larger organizations and teams. Every stakeholder may have a different view on the issue and have different opinions about priorities.

Data helps you to align all stakeholders to a common priority list.

How to document

You find ways to document downtime for OEE improvement in our article. We summarize briefly the main points below.

When to record

You can think about your machine in the following way:

• Idle: No Job
• Running: Producing
• Planned Stop: Maintenance, Breaks, Training, Setup/Product Change
• Unplanned Stop: Whenever something goes wrong

Important is to track every time the machine has a STOP both planned or unplanned.

Right at the machine, when a stop event happens, a journal entry should be made. The machine should not be restarted before the journal entry is finished.

Who should do it

Operators and technicians alike should write up any machine stops.

Operators should write up, If there is an interruption during production. Technicians should add to the log when they make maintenance.

What should be recorded

The record should include:

• Machine, Cell or Process-Step
• Kind of Stop (Planned, Unplanned) including reason (Maintenance, Break, …)
• Comment describing the problem / symptom.
• Comment describing the solution
• Time it took to resolve it

Example record

The log should be digital but can be in the simplest form Excel based.

You can use electronic shiftbooks or OEE tracker to have more support.

Peoplegeist offers a digital logging solution through the Wiki AI Production Kiosks

Analysis & take actions

Having digital documentation of your stops with failure description and solution is an important foundation.

2/3 of the work is to analyze the data and implement improvement steps.

Analysis steps

Production, maintenance and engineering should periodically review the biggest downtime reasons and the downtime trend.

Identify your biggest loss:

• Based on your digital log you can calculate the downtime per machine.
• Identify the machine with the biggest stop times.
• Group the losses by planned and unplanned stops.
• Use the comments to find biggest downtime causes

The comments help the team to find concrete issues.

If you have a lot of comments Text-AI can help you see repeating clusters and generate pareto charts for you automatically.

Standard strategies for dealing with issues

The problems listed are typically a mix of machine issues and operator issues. There are some standard strategies to address them:

• Unplanned cleaning / refilling / smearing / oiling /…
  These are regular maintenance actions and can be done as part of scheduled maintenance.
  • Add it to the maintenance schedule if it’s missing
  • Adjust the interval if it’s already on the schedule
  • Check if it’s done properly and reteach if necessary
• Wear and tear
  Parts that get worn out with use often have lead indications before they break.
  They make a sound or show visual signs of deterioration (surface not smooth, rubber becomes stiff and bridle, etc.)
  • Schedule regular inspections for those part to detect early sign of failure.
    Schedule replacement if early fatigue signs are detected.
  • See if there is a regularity behind replacement. This helps you to schedule the inspections or schedule replacements.
• Keep replacement part on stock and have instructions ready
  If the failing part fails repeatedly but rarely and without lead indication, you can have replacement parts ready. If it fails, it will be quick to fix.
• Start a root-cause analysis project
  Start investigating why it happens in the first place. This can be done regardless if one of the standard strategies fit.
  A simple method is the 5 Whys
• Gather more data
  If you don’t have enough detailed comments or data, Have a follow-up action for the whole team to write more detailed comments for this topic.

A root cause analysis can take some research and may take longer. Schedule follow-ups with appropriate people to have a quick 5 why session together:

• Operators who experienced the issues
• Engineers
• Maintenance technician who resolved it

Keep track of all the actions and decisions taken. Use for example a kanban board that is visible to everyone to document the actions and visualize the progress.

Meetings

The appropriate meetings will depend on your organization. This is a possible setup.

Your “Top Loss” Meeting should be well prepared. A person should be in charge of preparing it and making sure that the discussed actions are tracked and implemented. See more about that in the “Roles” section below.

It’s important that production and maintenance/technical service both take part. If you look at the “Standard strategies for dealing with issues”, it requires both teams to make changes. Both team need to be involved and have ownership in this meeting.

What you should consider for your meeting interval:

• Enough time to gather enough data: You need 30-50 records per assets to see recurring patterns
• Enough time for actions to be implemented

If you have a lot of stops, a weekly interval might be right for you.

You can also make top-loss meetings every 4 weeks with review window of 3 month.

Roles and Tasks

For this to work production and technical service need to work together. If one party is not actively engaged many change initiatives will fail.

The “Top Loss Master” is the person that drives the continuous improvement cycle forward.

• Has hands-on knowledge of the production process
• Is not directly involved in daily production (has time to analyze and think about the data)

A good fit for this role are:

• Production assistance
• CI manager
• Lean Manager
• Technical Lead / Production Lead

ideally this person used to be on the production floor but is now not tight to the daily grind anymore.

Summary: Bringing it all together

To successfully drive up your OEE and decrease your losses you need to:

• Document your losses with some minimum structure
• Production and technical service need to document together
• Loss improvement is owned by production and technical service together
• Have a “Top Loss Master” who can push through actions and changes despite daily production pressures.