Long time ago when we started with our CMMS, we defined priorities for maintenance work, from a maintenance perspective. We defined 4 levels:
1) Emergency (24 hours)
2) Urgent (48 hours)
3) Normal (2 wks)
4) Combined Shutdown

The same field has more values, but for maint workorders, only the 4 mentioned are active. The reason behind these 4 priorities was to avoid discussions about ASAP, very urgent, less urgent, maybe, next month or never

From our point of view, the time span indicates feedback to the client, not completion. But there are people (auditors) that insist otherwise. I can have an urgent situation at hands, that will take 3 weeks to complete.
An emergency in our eyes is, stop all other work now, all hands on deck.
I want to know how others have dealed with this situation.

In our case we avoid using the words emergency, urgent or asap because any works can become asap, urgent or emergency after they became prolonged backlog ie no actions taken or dragging their feets or the work process is slow or lengthy. So to avoid ambiguity, we define the work priority based on the consequence to set the response time as follows:

Priority 1 if there is already a HSE and/or production loss which shall be attended immediately. Work completion depends on complexity of the problem (as you rightly said above). If the problem is simple, then maybe it can be fixed within a short period of time but for complex problems, maybe temporary quick fix for restart up preferably after troubleshooting to identify the causes of failure and possibly followed by failure anaylsis, rca etc. Therefore, to avoid confusion, we make all work orders without the target completion date set by the system based on the response time & the user has to enter their own expected finish date before can proceed with work order planning. However, the response time to attend work notifications can still be measured and auditable by the time elapsed from the date of notification raised to the time of converting it into a work order with the end date to be set by the users themselves.

Priority 2 - potential HSE and/or production loss, to be attended with 24 hours eg for high vibration or other problems which cause shutdown or trip of vital or essential equipment without redundancy

Priority 3 - Normal maintenance to be attended within 3 weeks to allow for work order planning & scheduling including procurement of materials & services. The bulk of works should be under priority 3 so the planner should prioritize further into a, b or c to identify which works can possibly increase in risk = likehood x conseqence and thus their priority. Discussion with operations or technical services will help in these subpriorities.

Priority 4 - General maintenance for standing or noncore work orders, to be attended within 4 weeks.

Priority 5 - shutdown works especially planned turnaround.

Any works with initial priority 4, 3 & 2 can be become urgent or emergency if not attended until the last minute probably not knowing the consequences due to lack of knowledge (incompetency), insufficient information of purpose, lack of motivation, firefighting maintenance etc. Therefore, jobs for vital or essential equipment should be attended by competent personnel, also to ensure quality of workmanship.

The above definitions should be agreed by all departments, not only maintenance but also operations & technical services etc. One effective way to ensure people stick to the rules is whenever priorities 1 or 2 are raised, email will be auto-triggered to head of depts to inform them, then people will raise work notifications with the right priorities.

This message has been edited. Last edited by: Josh,

The way I see it there are two factors to be considered when defining priorities, the work type and the equipment type. I wrote something about this three years ago. I have attached a copy and I would be interested in your opinion on this.

Bryan Weir
http://www.pemms.co.uk

freedoc_Defining_Work_Order_Priorities.doc (34 Kb, 15 downloads) Defining WO Priorities

Bryan, Josh thanks for your input.

Bryan the matrix is usefull, but I have clients who's attitude is: "I don't care if the equipment has five bearings, or ten diodes, it is kaput, fix it!" Trying to explain the matrix to them is waste of time.

Josh in your definitions the words to be attended in how does it works in the real world? In my eyes it is not an issue , but the ISO audit clan is analyzing every comma on every document.

We do not set a target completion date either, so the priority is not time constrained.

What we do, the PPM jobs (always priority 3) will pop-up if after 2 weeks of release, no status change has occured. (the famous blacklist)

The e-mail notification, IT Dept still working on that sigh.., and the first item on the list is Priority 1, Emergency to the Plant Manager , not to give him a heart attack, but to catch the abuses in the system.

I think I need a Priority 5 for standing workorders.

"To be attended" - how does it works in the real world? This is a tough question to answer!

1) We have practiced this quite sometimes and the discipline engineers & managers have used to coming over weekends if need be. Any slack will be highligted by operations in the next all-dept morning meeting. So I think everybody tries to avoid the "heat" in the air-cond room there.
2) We set up a timetable of standby personnel to attend to any problmes which can lead to priorities 1 or 2 during outside office hours or night time, weekends & public holidays. That standby person shall not move far away from his resident or plant ie to arrive within 1 hours upon phone call if cannot settle via phone. There is also the timetable for store personnel. Operations shift superintendent (SS) can call discipline engineers or managers or storemen anytime outside office hours to clear any doubts on his side and get technical assistance, failing which it would be his failure to inform of the impending problems. Standby handphones provided foc.

3) If the audit is ISO9000 and the auditors go words by words, I think they still stick to the old idealogy of ISO 9001/2/3 which states "Do what you write and write what you do". Now after the latest ISO9000 revision in 2000, we just have to prove what we do as per our business process maps based on PDCA (Plan, Do, Check & Act) cycle. This is my view when I do audit as one of the ISO9000:2000 internal auditors. This means if you can prove to them that you attend to eg all priority emergency jobs with 24 hours in you case, then that's compliance. If there is a slight deviation, there could be an observation for your to rectify. But if there is a non-compliance eg you cannot attend to emergency jobs within 24 hours most of the time, then you have to improve the situation and analyse why is happening. If indeed the auditors check "every comma on every document", the auditors should attend the latest ISO lead auditor training or else only observation can be made for you to rectify linguistic errors like typos, grammars etc. When I'm the auditor, I only check for compliance and highlight matters which are probably not seen to the process owners or they already live with it so to guide them that it's not the best practice anymore. If the auditee agrees, then agree on whether it's an observation or non-compliance. However, I personnally believe zero non-compliance or no observations are contradictory to the spirit of ISO which subscribes to continuous improvement philosophy especially during internal ISO audits, we should detect all minor or any major issues as preparations to the external certification continuation audit. Smartly, we can use simple words to set the response time like "shall" which indicates must follow within the specified time or "will" which indicates a recommedation to follow. If the problem still persists, are you of one of the ISO internal auditors? All depts should have internal auditors to ensure "balance and check" and participate in the internal audit but audit other depts to avoid bias towards on own dept.

4) On the PPM blacklist, I think both of us agree that PPM should be carried out ontime or else we can set permitted deviations like -+10% of the completion duration, depending on local manpower or logistics and incorporate this in the ISO procedures. If you follow release date, there could be a problem if the release date is on the start date which I think not allowing for proper planning & scheduling with respect to other jobs at hands during that period of time. In fact, the PPM should be released 1 or 2 weeks earlier than the planned start date to allow for proper planning & scheduling to occur. In this case, it should be noted the PPM compliance is measured against the start date, not the release date. Normally we cannot adhere to PPM schedule because our all personnel busy with other corrective or incidental works in your case, right?. To avoid this, my machinery section has dedicated a few technicians to do PPM only while the rest to do incidental jobs. This arrangement always stays except instructed otherwise to cater for too many priority 1 or 2 jobs which are rarely occurring together. The result can be seen in the high PPM schedule-compliance for this section.

5) Upfront we track KPI for jobs with priorities 1 & 2 which indicate modes of maintenance ie a high figure here eg > 10% would indicates firefighting or reactive maintenance. So the aim is to minimize breakdown/onstream maintenance.

This message has been edited. Last edited by: Josh,