4th Condition Monitoring & Fault Diagnosis Conference March, 2010, Sharif University of Technology, Tehran, Iran

AN INVESTIGATION ON DEVELOPING MAINTENANCE PRACTICES FROM PREVENTIVE MAINTENANCE TO CONDITION BASED MAINTENANCE Case Study: Esfahan Petrochemical Company

ALI MOHAMMAD YASLIANI ESFAHAN PETROCHEMICAL COMPLEX

ABSTRACT
Over the past twenty years maintenance has changed, perhaps more so than any other management disciplines Moubray (1997:1). The changes in the oil plants are huger due to using high technology equipments. These equipments which enjoy complex design need new maintenance technique and skillful persons. Meanwhile Oil and Petrochemical plants are those industry that their maintenance cost are extremly huge. New maintenance techniques such as condition based maintenance and relaiability centered maintenance are intoduced to reduced maintenance cost by managing required maintenance tasks and resources as well as reducing cost of unplanned downtime. But where is the start point? where is the bundry of each techniques in practical? And finallly what is the benefit of condition based maintenance.

KEYWORDS: Maintenance Practices, Preventive, Condition based maintenance, Reliability Centered Maintenance

Maintenance planning manager MSC in construction management

4th Condition Monitoring & Fault Diagnosis Conference March, 2010, Sharif University of Technology, Tehran, Iran

Literature Search McAllister, Idhammar, Wilson and Willmott et al (in Wilson, 2002: 59-79) illustrate that maintenance practices and quality procedures are categorized in four main areas, which span both the organization and systems aspects of maintenance. 1. Activities and programs 2. People development 3. Planning and contractors 4. Control and computer systems The growth and change in development of maintenance practices can be said to be in three distinct stages, i.e. those in a period of sustained growth, which are being implemented or considered by many companies; those which are gaining momentum and are on the agenda for most companies; and those which are at a mature stage having once enjoyed popularity but are now waning.

Figure 1: Stage of development, Wave of change Source: Wilson 2002 Geraerds et al (in Coetzee Jasper L., 1999), the typical approach towards increasing the efficiency of the maintenance function is to implement some highly publicized philosophy or maintenance technique. These include reliability centered maintenance (RCM), total productive maintenance (TPM), condition based maintenance (CBM), computerized maintenance management systems (CMMS), auditing systems and the like. While each of these will certainly contribute to the success of the maintenance organization, the haphazard way in which they are introduced is a certain formula for sub-optimality. Coetzee Jasper L. et al (1999), the attention that the maintenance discipline receives is often of a haphazard nature without any proper integration between the various. The correct way of addressing the need for a very effective maintenance function in the organization is by having a more holistic view of the maintenance function. In this way the various philosophies, methodologies, techniques employed are properly coordinates and phased, leading to success. The effect of this is that a very successful maintenance improvement process for the organization is created. While each of the philosophies and techniques listed above play an important part in the solution, it must be implemented in a properly coordinated way. The problem with

4th Condition Monitoring & Fault Diagnosis Conference March, 2010, Sharif University of Technology, Tehran, Iran

the fragmented "solutions" towards improving the effectiveness of maintenance is not that each of the individual methods is not valid for what it intends to achieve, but rather that it does not assist the maintenance practitioner in positioning the technique as a part of a total maintenance strategy. The only exception to this is that of TPM which is a philosophy addressing the total complexity but which has had limited success in the western world due to a difference of managerial outlook Ingalls Preston et al (2009) in his article" Changing Maintenance Practices to achieve World Class Maintenance " on the TPM online website states that solid maintenance practices supports a strong maintenance system geared toward proactive activities involving the total organization. Improving those practices requires patience, management commitment and dedication, as well as the willingness to make it happen through well-conceived plans and actions. Best Maintenance has its foundation in Best Maintenance Practices. He strongly believes that one of the most important maintenance practices is Predictive Maintenance which has the following benefits: Lower maintenance costs Longer life cycle Less downtime Early problems identified Tim Henry, Geoff Baker, Tom Scott at el (in Wilson, 2002: 293) stay that maintenance of a plant unit based on its perceived condition is a more cost effective strategy than either maintenance on failure or planned preventive maintenance based on time or hours run. Maintenance on failure incurs interruption to production or services, more maintenance activity and extra spare usage. Planned preventive maintenance can results in unnecessary maintenance activity and downtime and excessive spare usage, whilst being ineffective in preventing breakdown. Wireman Terry et al (2005: 3-36) in his book " Developing Performance Indicators for Managing Maintenance" has proposed a pyramid which contains 11 elements, starting with Preventive Maintenance at the base of the pyramid, and working up to Continuous Improvement at the top.

Figure 2: Highlights a comprehensive maintenance/asset management strategy Source: Wireman Terry 2005

4th Condition Monitoring & Fault Diagnosis Conference March, 2010, Sharif University of Technology, Tehran, Iran

He believes to develop maintenance / asset strategy its need to start from preventive maintenance and then step to the other elements. It does not mean to start a new element in the higher stage; the entire element in the lower stage should be completed. But its need the relevant infrastructure in the lower stager has been prepared. In this study a hard effort is made to trace EPC,s activities and program from preventive maintenance to the condition monitoring and reliability centered maintenance and compare it with the stage of developments shown in figures 1 and 2. EPC maintenance history As soon as construction phase is finished maintenance department was organized. It employed project peoples who were interested and capable in the maintenance. They started to repair failures and return down equipments to the production. After that time during the years following improvements were happened: Preventive Maintenance In EPC, Preventive maintenance was started by defining certain routines based on experience or lessons learned from other petrochemical complexes at 1993. Many check lists for equipments were prepared .Spare parts for most of the equipments identified and the relevant minimum and maximum were calculated. Two years later Calibration was added to the PM. Scheduled preventive calibration, inspection, services and checkups are performed to maintain EPC production equipment and facilities at the best possible operational status and to prevent them from unexpected failure during normal operation. All rotary equipment had the following routines: Routine A: an overall check of the equipments Routine B: Equipment should be stopped to do some detail checks such as alignments and oil replacement Routine C: The equipment should be stopped, disassembled and all parts will be inspected. The damage parts or parts which are recommended by the manufacture were replaced. The last mentioned routine cost yearly about 1480 million Rails for EPC. (Before implementing Condition Monitoring System) Meanwhile based on the manufacturer's recommendations comprehensive overhaul performed once every two or three years, which includes service, repair and parts replacement on those equipments that cannot be shut down for routine services during normal production. In each oil and petrochemical plants overhaul is the most expensive maintenance activities. A huge amount of spare parts will be consumed, plants should be stopped and we loose production. Predictive maintenance Condition monitoring was started by a simple overall vibration measuring at 1997. It completed by adding sound and oil analyses. The main questions at that time were Which equipment should be considered at condition monitoring system? Which technique is more suitable or cost effective? What is the optimum interval for collecting data and analyzing? What is the relation between preventive maintenance and predictive maintenance or where is the boundary.

4th Condition Monitoring & Fault Diagnosis Conference March, 2010, Sharif University of Technology, Tehran, Iran

After many try and error methods EPC used the priority equipment model to select equipments that need to be in CM. EPC has categorized the equipments in the following four main groups. This categorization is mostly made based on equipment's impact on production (Continuous or capacity) and maintenance costs. Recently safety and environment aspects are considered as complementary elements to review and revise the current equipment categorization list. Vital Equipments (n=4) In EPC, vital machinery or equipments are those that must be online for continued plant operation. Loss of any one of these components will result in a plant outage and total loss of production. This group includes equipments which are single (without no spare or redundant) in the process line and any equipments breakdown means production shutdown. Very important Equipments: (n=9) In EPC the following equipments are put in this group: 1. This group covers such equipments which have redundant and most of the time only one of them is in services. 2. Equipments that their break is not equal to production break because it has redundant but its failure would increase risk of loss availability for plant. 3. Equipment which its break equals to reduce production capacity 4. Equipment which its break equals to certain numbers of other equipments shutdown. Important Equipments (n=36) In EPC, this group covers such equipments which have redundant and most of the time only one of them is in services. But important equipments or machinery include major plant's equipments that does not have a direct (online) dramatic impact on production but they are used to support systems in emergency situation Other Equipments (about 15953) Other plant equipment that has no serious impact on production or the relevant maintenance cost is not so much. Equipment prioritizing has a process in EPC and when new equipment is added to the company's assets a special form should be completed to identify the proper priority by the different skills. Based on the above categorization all vital, very important and important equipments are directly under condition monitoring consideration. Vibration, oil and sound analysis are done with the 15 -30days interval. A detail study will be performed for any breakdown on these equipments sometimes it redound to use other techniques such as thermograph. For the other rotary equipments which have continues services but are not in the first three group only vibration analysis with 30 to 60 day interval is planned. Using the mentioned condition monitoring techniques helped EPC to reduce time based routines and increase condition based maintenance. This happened by postpone routine C of the equipments. The benefit of this change is saving about 2581 million Rails yearly. This is only direct cost saving which comes from resources saving such as spare parts, man-hours and tools. To calculate correct cost saving amount we should consider all indirect cost saving amount such as down time cost. The last and perhaps the most important question is that "Where is the boundary of the condition base maintenance". How time a recommended routine could be

4th Condition Monitoring & Fault Diagnosis Conference March, 2010, Sharif University of Technology, Tehran, Iran

postponed. The author of this paper believes that to answer the question the following points should be considered: 1. Costs a. The condition monitoring costs vs. fix time routines cost b. The condition monitoring costs vs. emergency downtime cost c. The condition monitoring costs vs. early failure finding benefits 2. Abilities to detects all deteriorations Do you think a simple vibration and oil analysis could detect all deterioration? Vibration, oil, and sound analysis which are most popular in Iranian oil companies could mostly detect rotary equipments failure. But certain huge equipments such as steam generators have other important parts. So when a recommended fix time routine based on the some (not enough) condition data will be postponed (Omitted) these parts would be missed from inspection. For example, EPC has three steam turbines (Generator), after many time which the relevant C (a detail inspection and changing defected parts) routine were postponed , EPC has decided to do routine C based on an unknown vibration peak. We balanced rotor to solve this problem. But beside that we found many erosion and corrosion on the deferent part of the turbine as showmen in the following figures. This is more serious for subsidiary equipments of main equipment.

Figure 3: Steam Turbine Damage Parts Source EPC, 2009 Based on the above, it could be concluded that a condition based maintenance versa fixed time routine is limited to the amount and level of the condition monitoring. To rely on the condition based maintenance you need to trace all type of the conditions to detect any deterioration at the early time.

4th Condition Monitoring & Fault Diagnosis Conference March, 2010, Sharif University of Technology, Tehran, Iran

To prevent any damage which could not be detected by the existing condition monitoring system, EPC has decided to investigate the postponing routine C of the rotary equipment through an expert and skillful team. The input of this expert team is all technical information, maintenance data and the relevant maintenance cost. The recent studies show that RCM is a solution to define boundaries of the maintenance practices from run to failure, preventive and predictive techniques. EPC has measured certain maintenance indicators and categorized them in three different group costs, equipments performance and systems effectiveness. The following indicators prove that EPC has implemented proper maintenance practices from preventive to condition based maintenance. Availability for the Vital, very important and important equipments The following figure 3 shows that EPC has improved his equipments with the priority of vital, very important and important from 98.58% at 2006 to near 99.96% at 2008.
100.50% 100.00% 99.50% 99.00% 98.50% 98.00% 97.50% 2006 2007 2008 2009
2009

Year

AVAILABILITY

Traget

Figure 3: EPCs availability indicator Source EPC, 2009 No of failures for the Vital, very important and important equipments: The following Figure 4 illustrate that EPC has reduced prioritized equipments failure. The world class is zero failures.
6.00

5.00

4.00

3.00

2.00

1.00

0.00 2006 2007 2008

Year

No of Failures

Traget

Figure 4: Number of Failures Source EPC, 2009

4th Condition Monitoring & Fault Diagnosis Conference March, 2010, Sharif University of Technology, Tehran, Iran

Break Down work orders ratio This indicator demonstrates the ratio of the break down work orders man-hours to the total work orders man-hours. Break down work orders are those works that perform after a functional failure of the equipments. These failures have down equipment. The world class is zero. Figure 5 shows the amount of break down work orders are is reduced from 3.09 % at 2006 to 1.60 % at 2008.

10.00% 9.00% 8.00% 7.00% 6.00% 5.00% 4.00% 3.00% 2.00% 1.00% 0.00%
2.00% 3.09% 1.80% 1.83% 1.60%

1.35%

2006

2007
Emergency Workorders Ratio Target

2008

Figure 5: Emergency work orders ratio Source EPC, 2009 Preventive maintenance ratio Following figures 6 and 7 present the number of the preventive maintenance work orders and the relevant man hours respectively. The trends show although EPC has increased his preventive maintenance jobs but the amount of the consumed man-hours is reduced.
70.00%

65.00%

60.00%

55.00%

50.00%

45.00%

40.00%

35.00%

30.00%

2006 PM Ratio (No)

2007 Target

2008

Year

Figure 6: Preventive maintenance ratio (No of PM) Source EPC, 2009

4th Condition Monitoring & Fault Diagnosis Conference March, 2010, Sharif University of Technology, Tehran, Iran

80.00%

70.00%

60.00%

50.00%

40.00%

30.00%

20.00% 2006 2007 PM Ratio (Man Hours) Target 2008 Year

Figure 7: Preventive maintenance ratio (Man hours) Source EPC, 2009 Spare parts (replaced parts) cost to maintenance total cost Following figure 8 prove that EPC has reduced his spare parts replacement cost ratio during last three years.

32.0% 31.0% 30.0% 29.0% 28.0% 27.0% 26.0% 25.0% 24.0% 2006 2007 2008 26.5% 26.8% 31.4%

Figure 8: Preventive maintenance ratio (Man hours) Source EPC, 2009 Conclusion: EPC has employed the integrated maintenance practices based on maintenance objectives and practices maturity status described in figure 1. EPC has set up its maintenance / asset strategy on the preventive maintenance techniques and developed it by implementing; predictive maintenance and CMMS computerized maintenance. Based on the figure 2 as the needed infrastructures are prepared EPC could start RCM. The author of this study believe that EPC is a sample of the Iranian oil, gas and petrochemical complexes and as the most work flows, disciplines in the mentioned

4th Condition Monitoring & Fault Diagnosis Conference March, 2010, Sharif University of Technology, Tehran, Iran

companies are the same (mostly), experience would be useful for them. References:

all findings of this study

and mentioned

1. COETZEE J. L., "A holistic approach to the maintenance "problem"". Journal of Quality in Maintenance Engineering, Vol. 5, p. 276 Retrieved May 24, 2009, from ABI/INFORM Global database. (Document ID: 86926405). 2. INGALLS, P., Changing Maintenance Practices to achieve World Class Maintenance 2009, available from: http://www.tpmonline.com/articles_on_total_productive_maintenance/tpm/newpr actices.htm 3. MOUBRAY, J., Reliability centered maintenance, Second edition, Butterworth Heinemann UK, 1997 4. WILSON, A., Asset maintenance managemen, first edition, Industrial Press Inc., USA, 2002 5. WIREMAN, T., Developing Performance Indicators for Managing Maintenance, Industrial Press Inc., New York, 2005

6. YASLIANI, A.M. "An Investigation on Development Maintenance Strategies in Iranian Petrochemical Companies", submitted dissertation to GGSB and IUT Universities for Degree of Master of Science, 2009