There is no Planning and Scheduling Standard
There is no Planning and Scheduling Standard but there are Lots of Books
Note: This article focuses on weekly maintenance scheduling but for more information on maintenance planning, see Maintenance Planner Duties .
Asset-intensive organizations should have a good grasp of asset management so that they can promote asset reliability, workforce productivity, and job safety. That said, there is no documented P&S standard, which is unfortunate because most agree weekly maintenance scheduling is an industry best practice. Further, it is quite common to have more maintenance work than available craft resources, meaning it is extremely important that leadership selects the right work each week.
Smaller organizations are less likely to have adequate numbers of planning/scheduling staff and therefore decide not to pursue weekly scheduling. But, if they had an automated solution within the CMMS, they could. For others, if the selected work was subjectively based (i.e., who yells the loudest), then they risk performing work that is not as essential. By implementing an automated scheduling process based on risk-ranking, you are gaining speed and accuracy which enables you to re-run as necessary up until the last moment.
What are the Benefits of a Weekly Schedule?
Without a weekly schedule you can’t measure schedule compliance; nor can you identify break-in work. There are two ways to define reactive maintenance: (1) work that requires immediate attention such as emergency/urgent, and (2) unplanned work which is executed but not scheduled, i.e., outside of the weekly schedule.?
Vendors are Redefining the Weekly Schedule Definition to fit their Proprietary Software
As an industry best practice, a more practical approach is needed which speaks the truth. I am basing my definition of a weekly maintenance schedule on recognized authors of planning/scheduling books - which is simply a "set of work" for the week. Too many vendors have twisted the definition of weekly scheduling. They have created graphical tools that are click-count heavy but lack automation. Further they have skipped the weekly schedule step and gone straight to building 5 daily plans. The odds are great these daily plans will need adjustment due to emergent work and slippage of previous days - which is unnecessary rework.
Weekly Maintenance Scheduling - the Ideal Process
With resource leveling built into the CMMS, the planners use this initial pass (which contains rough estimate plus fully planned jobs) to create work packages. After 3 to 5 days, this planned work represents the optimum level of work for the week with the greatest risk to operations and that work which maximizes craft utilization. Modifications, such as opportunistic scheduling, can be made however up until the last moment. This set of work also forms the baseline from which compliance is measured.
Although advanced processes have many prerequisites, they also provide the greatest payback in terms of efficiency and productivity. But without alignment of resources (software, data, roles), weekly maintenance scheduling can be near impossible. For example, the backlog needs to contain both repair and PM/PdM activities. The backlog also needs a relative ranking of importance. Automated leveling needs craft estimates, craft availability and craft efficiency values. Carryover work from last week must also be captured, called ETC, which must be used instead of the original estimate. As to role requirements, a gatekeeper should review all incoming work and provide an immediate rough estimate for plannable work.
Note that for very small organizations, most do not have planner/scheduler positions. Therefore, anyone could be given this responsibility for reviewing incoming work and assigning a rough estimate (with lead craft). Example staff include maintenance manager, maintenance superintendent, maintenance supervisor, maintenance engineer, or anyone familiar with maintenance and has a fair amount of seniority.
Bad Data However Severely Impacts Ability to Automate?
Whether you are doing failure analysis or automated scheduling, bad data will stop you in its tracks. For some, their number one issue is bad schedule data. Most CMMS products do not even provide a field for ETC (by craft), or a craft efficiency field. More importantly, they lack a resource leveling algorithm to prevent over (or under) scheduling of craft resources. Having bad data may be the number one reason why organizations prefer to manually create schedules but fear of automation is a close second.
The best-in-class organizations will pursue advanced processes so that they can (1) minimize risk to operation; (2) maximize craft utilization; and (3) optimize backlog reduction.??
All P&S Books Describe a Manual Process
This is where the planning/scheduling book authors fall short, as they typically end up describing a manually created schedule. Another word for manual is subjective (or best guess). Pretend there is a backlog of 500 or more work orders and you have a (weak) 1 thru 5 priority system. This is where the craft supervisor chooses work from the backlog (some planned and some not planned). His next step is to create a daily plan where the worker is assigned. The add-on scheduling products perform subtraction from availability - which they call "automation".
Because the above process can be fairly intensive (i.e., a lot of clicking), very few organizations have the staffing to produce a true weekly schedule, with or without add-on software. With intelligent automation, any size organization could produce this output. True automation means a single person can run a leveling program and generate a risk-based, weekly schedule .... in 1-2 clicks.
Users are Searching for a Workable Solution
Planning and scheduling has always been a hot topic as seminars typically have hundreds of attendees. There are many planning/scheduling books but if you read closely you will note that the topics are primarily about planning. In the case of conference venues and virtual seminars, the speakers often prescribe a solution that is based solely on their software solution, and ignore the overall footprint requirements (planning and scheduling staff man-hours) to manage the data, critical roles and process to create this output.
It is my belief that the attendees are continually looking for a powerful but workable solution which fits their staffing. For example, a maintenance scheduler shouldn't have to sit with each supervisor for 1-2 hours each week building 5 daily plans (and skipping the weekly schedule step).
Use Intelligent Automation to Create Multiple Passes
Keep in mind that there is no human that can evaluate hundreds (or thousands) of work orders in a backlog with multi-craft requirements and come up with the ideal set of work, all in their head. With that understanding, why not let the software do the resource leveling, and then, have leadership perform a secondary review (e.g., in the weekly schedule oversight meeting)?
Refining the Set of Work
A more practical approach would be to establish a "set of work", as a first-pass, out of the CMMS using a resource-leveling program (RLP). The initial set of work is used by the planners to create formal work packages. Additional resource leveling can be performed in the days prior to, and during, the weekly schedule meeting to optimize the scope and craft resources. Once issued, each craft supervisor would reference the approved weekly schedule to create the daily plan before each shift - and incorporate any emergent work from last 24 hours. The Daily Plan would also be a report generated from the CMMS and show work for this lead craft as well as support craft role. Sadly, the majority of all industry sites continue to struggle in this area – and some have given up. The internet may be abuzz with AI and IIOT promises but no one is talking about automation of maintenance scheduling.
Prerequisites to Automation
When evaluating reasons why companies DO NOT create a weekly schedule the number one reason is their bad data. This problem can be fixed however once you implement an improvement plan. In addition to bad data, there is a fear of automation. The CMMS should be allowed to create a first pass at creating the initial set of work that will be used in the weekly schedule.
领英推荐
Here are 6 actions to configure your system:
(1) Define "bad data"; describe methods to fix bad data.
(2) Create a Data Quality Plan that describes responsibilities, critical fields essential for this advanced process, periodic audits, proactive error checks, a flow chart to the endgame, and training.
(3) Implement new role, such as Gatekeeper.
(4) Implement new fields to capture critical data such as Estimate to Complete, rough estimates, and craft availability (with efficiency factors).
(5) Implement resource leveling algorithm inside the CMMS; implement risk-ranking matrix; implement Order-of-Fire application.
(6) Alter processes in support of the above.
Operations Must do Their Part
The service request must receive a "level of importance" designation. Examples are emergent work such as Priority-1 as emergency, and Priority-2 as urgent. This work is dispatched once converted to a work order. If not emergent then priority defaults to 3 - which is plannable work. The maintenance staff (such as gatekeeper role) would evaluate the priority-3 and categorize as 3,4,5,6,7 to provide rough backlog prioritization.
WO Priorities as a Whole are not Adequate to Rank the Backlog
Create a 2-dimensional risk matrix with categories of work across the top. Create a program that is fired nightly to apply a numerical ranking to the entire backlog. It is this ranking that the resource leveling program uses to process the backlog - not the WO priority.
Majority of Backlog is Missing Estimates
Recognize the importance of a rough estimate (lead craft and rough estimate applied by gatekeeper role). Determine where this rough estimate is stored on the work order application. The rough estimate would be superseded by the formal estimate once entered. Once this new process, is started, at some point the entire backlog will have an estimate. For resource leveling to work, every work order must have an estimate.
No One is Available to Apply Rough (or Formal) Estimate
Require the requester (operations or student) to enter probable lead craft (which then crosses over to the work order). Create subroutine on work order application to auto-calculate a rough estimate. Once a year, run program to establish average (actual) hours by lead craft and store this inside the CMMS. This value is then used to auto-populate the new work order.
No ETC Estimates are being Captured
Add field to craft estimate line to store this overriding value. Make sure leveling program uses this value if entered. Train maintenance supervisors/techs to enter this for every INPRG work order (if at end of week) - called carryover work.
Book is Targeted for Summer 2023
I’ve been writing a book on weekly maintenance scheduling for some time but it’s been hard for me to finalize my thoughts – until now. I think the post picture showing an outline in WBS format will help organize the key points and make it easier for the reader to understand.
Below are some of the many posts on weekly maintenance scheduling I've made over the years [links shown below] .
A link defining the history of C/SCSC which was a DoD directive [not a standard].
A list of ISO standards can be found here .
Specialist in Asset Management, Maintenance Management, Inventory, Procurement and Supply Chain. Expert on IBM Maximo
1 年Good article on a very important subject. The crux of the matter lies in good quality Metadata capture on Asset, PM and Job plans. In my experience, weekly scheduling has been an area which has not been given traction by Maintenance Leadership. It is something Maintenance really need but don’t know how to get it. Common Sense approach is indeed the way to go. Building trust on the data supports using the CMMS.
Maintenance engineering. Modern CMMSs. Solutions to chronic downtime issues. Asset and maintenance management advisor/consultant
1 年The sentiments are valid. I have been in the maintenance delivery arena for decades. Each situation calls for its approach. Scheduling for 40 haul trucks running 24x7 maintained by crews on 2 week rotating required a different approach to a foundry on 5+2, with most of maintenance on weekends. In each situation, we spent time developing the system. Then did what was necessary to make the CMMS support the system.
Planificación / Gestión de Activos / Confiabilidad
1 年As always John, you made us think and rethink some ideas, which we took for granted. I often say that this is easier said than done, and this is especially applicable in this situation, because sometimes we assume that scheduling is a simple "list of activities"... but that is very far from the truth! That "holistic approach" that you mention is a very good way to help reach an agreement with the technical staff to achieve the programs and, thanks to that, achieve the plans. #Backtotheroots #Reliability #Confiabilidad #Plantaconfiable
(BERA Member) NASEN Member. PATOSS Member. Neurodiverse Youth SEND & STEAM Education IAG. Catering chef Transition and Career Development. Ed.CMS. CRL&CMM Eng C&G TAQA. Work-based educator. Instructional Design.
1 年A scaffold is a useful idea that can be used to build on an idea and culture as a point of reference in an ecosystem. Ideally, this culture should be evidence based and be fit for purpose in my humble opinion.
Taylor Reliability Consulting
1 年Scheduling of work in any maintenance organization that I encounter is either not done or it is done very poorly. Most CMMS systems are not equipped with a good work scheduling module and that leaves a void for the generation of a decent schedule. And as stated, many organizations are happy with just a simple list of work to be completed sometime during the week. I have an example of a list of work to be completed sometime during the year. Good work management as it relates to scheduling is a big gap in industry for the companies I encounter. And there certainly is no Scheduling Standard. Planning is much more straightforward than scheduling and a planning standard could be easily drafted.