The Problem with Weekly Maintenance Scheduling

Most organizations agree that weekly scheduling is an industry best practice. That said, most organizations do not have one. Why is this? Is it a software issue or a data issue? Perhaps they have inadequate P&S staff. Is there simply a lack of understanding as to the benefits? What are the requirements to have a weekly schedule? Maybe leadership does not believe they need one or see the benefit. Maybe they prefer reactive maintenance.

The Definition

A weekly maintenance schedule is a “set of work” for a week, identified in advance. This work scope is ideally resource-balanced such that the total craft estimates are equal to craft availability for one week. As the week begins, the daily plans are extracted from the weekly schedule.

Some believe a weekly schedule is 5 daily plans linked to the day of the week and immediately given worker assignments. This process leads to a lot of rework (re-editing the daily plans) because there will always be emergent work, and these plans need to reflect what is going to happen that day. Plus, there is day-to-day carryover work that needs to be reflected on the following day.

The Weekly Maintenance Schedule Process

There are two approaches to weekly scheduling – manual and automated. The manual process involves a lot of (a) manual calculation, or (b) plain guess work. Sometimes, they use the same count as last week to determine how much work they can do. The weekly schedule review meeting simply becomes a cursory approval.

Automated resource leveling is more accurate because it mathematically compares craft estimates to availability and maximizes craft utilization. The weekly schedule review meeting allows for last minute changes, what-if analysis, and work order bundling (called opportunistic scheduling). The use of automated tools is especially beneficial to smaller organizations who may or may not have adequate planner/scheduler staff.

The Benefit of a Weekly Maintenance Schedule

If you were fortunate enough to have a weekly schedule, these are the benefits:

A Comprehensive P&S Approach is needed to Reach Optimum Levels of Uptime

In the 1980s DuPont commissioned the largest ever benchmarking study of maintenance and reliability practices. This study included of 3500 sites across North America, Europe, and Japan. The study concluded that 5% of these companies, the so-called “Best of the Best”, do planning, scheduling, PM, condition monitoring, and defect elimination.

The Planning, Scheduling, and the Defect Elimination Process

Planning and scheduling, when done right, support reliability. O&M feedback supports continuous improvement of the planning process and maintenance strategies. A weekly schedule process supports proactive maintenance, reduces delays, and promotes job safety. With intelligent automation, the scheduling process is simplified and made more efficient.

It is also possible to apply craft (rough) estimates to all unplanned work, globally, in the backlog by evaluating actual labor hours by craft as a one-time action.

Key Roles in the Process

REQUESTER creates service (work) requests. The requester clearly states the problem (description, asset, location), and assigns the proper level of urgency, i.e. 1 (emergency), 2 (urgent), and 3 (plannable). The gatekeeper determines where the plannable work falls in the backlog.

GATEKEEPER role triages incoming work. Emergency/urgent work is dispatched. All other work is deemed plannable. Note: some organizations have a 24/7 call center in lieu of a gatekeeper. This role is extremely important to qualifying and estimating the backlog.

Additional actions by the gatekeeper include assess feasibility, verify not duplicate, enter basic priority (3 through 7), enter asset problem code (i.e., pump will not start), indicate if functional failure (Y/N) has occurred, work categorization, and flags HSE/risk if appropriate. He then changes status to waiting planning. Work categorization values include design flaw, supv force, asset condition, operability, maintainability, energy-efficiency, and ergonomics.

CORE TEAM establishes a risk matrix. This is a one-time effort but can be altered over time.

Note that equipment criticality is used as the horizontal reference, except for the last two verticals, which use work order priority.

Notice how the vertical categorization columns affect the numerical ranking. But if HSE/Risk is applied to the work order, then this becomes the dominant factor.

HSE/RISK MANAGER reviews any work routed to him and grades accordingly using a hazard matrix. The resulting value may dictate his oversight once the job is scheduled.

Automated Ranking of the Backlog

CMMS CRON TASK (re-)ranks the entire backlog using a prebuilt risk matrix each night.

MAINTENANCE SUPERVISOR updates craft availability weekly (inside the CMMS) – including planned absences. This needs to be accurate just before calculating the weekly schedule.

RESOURCE LEVELING PROGRAM (or RLP) resides inside the CMMS. The gatekeeper (or scheduler) runs this leveling program once a week. This creates a “set of work” which includes work orders that are fully planned or with rough estimates. The planners would then finish planning the rough estimates to establish a fully planned "set of work".

PLANNERS create work packages that identify craft/skill requirements, materials, special tools/equipment, safety plans, and possible services required to do a job. A complete work package might also include references to drawings, procedures, OEM documentation, shutdown requirements, and access or lifting requirements. They then change status to fully planned (or waiting material).

GATEKEEPER RUNS RLP A 2ND TIME: The gatekeeper (or scheduler) runs the RLP to create a preliminary weekly schedule. Process order is determined by an “order of fire” application containing multiple SQL select statements.

WEEKLY SCHEDULE MEETING: Representatives from operations, maintenance, planning, engineering, and HSE/risk management will be present. They will review the computer-generated weekly maintenance schedule. They may decide to perform work order bundling (bringing in lower priority work), add new work, or drop others. In any case, this group has the final say over the weekly schedule work scope. They run the RLP a final time to create the official schedule which is frozen for schedule compliance.

The Automation Timeline

It is quite common for more work to be received than planners can plan, which is okay, but how do they know what work to plan first from an ever-growing backlog?

This is where a carefully designed ranking matrix is most helpful that tells the RLP what to focus on first. Thus, the process would be RANK THE BACKLOG – RESOURCE LEVEL – PLAN THE SET OF WORK – REVIEW/APPROVE SCHEDULE. The set of work is created 3 times as shown in steps (5), (7), and (9).

Role of the Maintenance Supervisor

It is true that some work going into the weekly schedule may not be fully planned. This happens when there are insufficient work packages ready for execution. A second reason for this happening is that a high-priority job was just created and there was no time for planning.

Therefore, in these instances, the maintenance supervisor will have to perform field planning. But the first step is to create a daily plan. They must select their craft work from the weekly schedule for Monday and link to the worker's name. They also must accommodate any emergent that comes up in the last 24 hours. They repeat this process before the start of each shift.

Other duties of the maintenance supervisor are:

• Build the daily plan; notify operations of upcoming work needing permits
• Discuss the daily plan with the crew; hand out work orders
• Conduct JSA/JHA review; encourage sharing of past experiences including hazards
• Visit each work site during the shift to address issues and get expectations of completion. If job is expected to carry over to the next day, then he needs an ETC (estimate to complete).
• Start building daily plan for next day.
• For those carry-over jobs, the supervisor should provide shift turnover notes.

Role of the Job Site Leader

When the job is done, the CMMS is updated. Actual hours are entered no matter what at the end of shift. Either the job lead or supervisor must enter the ETC if work carries over. If the job is 100% done, then the status becomes complete, actual hours are updated, failure codes are entered, and feedback is provided.

Weekly Schedule (WS) Compliance is to the Week, not the Day

If you can't measure compliance, the odds are you do not have a true weekly schedule process. WS Compliance picks up scheduled work that was not started, and, any work done by staff that was not on the weekly schedule or daily plan. In other words, this work was performed without permission and requires an explanation.

The Resource Leveling Process

The Order of Fire Application Runs the Resource Leveling Program (RLP)

This application establishes the processing order using SQL statements and is totally controlled by the core team (or CMMS admin). Running the RLP also sets the week start date. Some work orders have date fields which determine if this work “fits into” the scheduled week. That is the purpose of the control date.

This is an Advanced Process

Advanced processes are the backbone of high-reliability organizations (HRO). They also require CMMS configuration. Automated weekly scheduling has a lot of working parts, but once in place, the CMMS can generate this output in 2-3 clicks -- which is the definition of “user friendly.”

Currently, only large organizations with planning/scheduling departments can generate a weekly schedule, but what if there was a program that resided inside the CMMS that performed resource leveling? How can we overcome any challenges?