Optimum Main Equipment Replacement Time for the Power Plant in the Port Harcourt Refinery
Chapter One
Objectives of the study
The major thrust of this research work is to critically examine the existing maintenance procedures/culture in the new Port Harcourt refinery as a basis for proffering solutions to the problems of low refinery products output. The work will ultimately lead to the presentation of a mathematical model for use in determination of the availability of the plant machines. Effort would be made to use the result of the availability analysis and other existing models in the literature to prescribe a replacement period for the critical equipment.
CHAPTER TWO
LITERATURE REVIEW
Analytical techniques provide the means whereby we get a closer insight into the ways in which a service is implemented. Applying a variety of analytical techniques ensures that results are corroborated from all angles. These will indicate how to deal with new and emerging maintenance problems, Priel (1974). A great many analytical techniques that lie within the field of plant maintenance have been developed in recent years, and it is probably only a lack of time, resources and good understanding that prevent crude oil refineries from applying all of them profitably to maintenance operations. Most of the analytical techniques of maintenance in the refinery deals with inspection strategies, over haul policies, preventive maintenance planning, Replacement, size of service crew; and corrective maintenance. All these topics ought to be treated in a way towards finding optimum economic solutions by means of mathematical equations.
Inspection Models for Maintenance
Inspection is an essential part of the maintenance program in the crude oil refinery. Many machines require some form of inspection to provide verification and confidence that the system or certain combination of equipment is performing its intended function. Inspection for the refinery maintenance can be classified as either internal or external. Internal inspection implies the watching of internal parts (such as gears, bushes, bearings, tolerances in parts e.t.c.) for possible failure when the refinery machines are under planned shutdown.
External inspection is basically the same thing as monitoring. It tends to detect abnormal sound, vibration, heat, smoke, etc. when machines are in operation. The formulation of inspection models is usually based on the fact that the frequency of inspection should be decided very carefully, as too few inspections may cause breakdown, as defects could not be traced out and rectified immediately. While too many inspections means wastages of machine time and Labour productivity. Usually, inspection models determine optimal inspection schedule by delays in failure detections. Generally, in an inspection model, the decision is usually to determine the sequence of inspection times:
(x = x1, x2, …xn) that minimizes the expected cost per cycle or the expected cost per unit time.
Balow et al. (1963) is a reference point for a majority of machines inspection procedure. The authors gave the total cost per inspection cycle as:
C1 (t: x) = C1n + C2 (Xn – t) (2.1)
Where
C1 = Cost of a single inspection.
C2 = Cost of leaving an undetected failure per unit time
n = expected number of inspections per cycle.
t = time to failure.
X = (x1, x2 …) which is the sequence of inspection time and n is such that
Xn-1 < t £ tn, and x0 = 0.
The model minimizes an optimal policy to be that which minimizes E {C1, (T, X)}
Applying the necessary conditions
¶/¶ x k E {C1 (CT, X)} = 0 gives
X k + 1 = Xk +
K = 1, 2, 3…….
So that the optimal policy (X1, X2,…Xn) is specified once the optimal X1 is known.
F(x) and f (x) are the systems life time reliability and density function respectively.
Overhaul Policies.
The overhaul policies in the new Port Harcourt refinery are usually to shutdown all machines at two years intervals and carry out a turnaround maintenance (TAM) on all machines. This method however seems stereotyped as little regard is paid to the realities of such policy. Much as it attempts to renew all machines, the maintenance interval is rather too long and therefore cannot reveal the need to replace a particular machine, or even trace an unsuitable component. This method also lacks the indices of control which could indicate whether breakdowns are running wild or not. It encourages carrying out both corrective and preventive maintenance on all machines and hence evaluating the actual cost of the operation becomes difficult.
CHAPTER THREE
METHODOLOGY
OVERVIEW OF THE NEW PORTHARCOURT REFINERY
The plant used for the study is the Port Harcourt Refinery Company Limited (PHRC) located in Rivers State of Nigeria. The plant processes crude oil. It is one of the subsidiary companies of the Nigerian National Petroleum Corporation (NNPC).
Within the refinery site, there are two separate plants, old and new plants. The old plant was commissioned in 1965 while the new plant was commissioned in 1989. The two plants have a combined designed refining capacity of 210,000 barrels of crude per stream day (BPSD). The old plant processes 60,000 BPSD while the new plant has an installed capacity of 150,000 BPSD. The main functions of the refinery are to receive crude oil from Shell Petroleum Development Company field at Bonny via the Pipelines and Products Marketing Company (PPMC) limited, another subsidiary of NNPC and process the crude oil into useable products such as premium motor spirit (PMS), automotive gas oil (AGO), kerosene (house hold and aviation fuel), liquefied petroleum gas (LPG), fuel oil, etc. The major process units of the PHRC are situated in five areas: areas 1-5. The new refinery is made up of areas 1-4 while the old refinery is area 5. Area 1 has the following units: Crude Distillation Unit (CDU) and Vacuum Distillation Unit (VDU). Area 2 consists of Naphtha Hydro-treating Unit (NHU), where naphtha is hydro-de-sulphurised; Catalytic Reforming Unit (CRU), responsible for upgrading naphtha to higher octane value reformate; Kero Hydro-treating Unit (KHU) where kerosene is treated to make it acceptable for aviation use; Continuous catalyst Regeneration Unit, which constantly reactivates the deactivated catalyst from the reformer.
CHAPTER FOUR
ANALYSIS OF MAINTENANCE DATA
Preamble
The maintenance data used in this study were obtained from the Planning and Budget Monitoring Department of the PHRC and by personal interview with the management team and key operating personnel of the PHRC.
The maintenance data obtained included the following:
- Process units down time in hours for five years (2000-2004).
- Power plant and utilities down time in hours for five years (2000-2004)
- Frequency of process units, power plant and utilities down time for five years (2000-2004)
Below were the above data in tabular forms and in bar chart plots.
CHAPTER FIVE
RESULTS AND DISCUSSION
Equipment Replacement Policy
It has been shown that the major breakdown occurred at the turbo-generating unit. The availability of 38.1% to 62.1% during the years under reference (2000 to 2004) shows an operational availability decline which varies from 37.9% to a maximum of 61.9%. This of course is expected to affect the productivity of the refinery negatively. This essentially calls for the determination of the replacement period of the turbo-generators. As mentioned in chapter 2, Sharma (2005) suggested that when time is measured in discrete units as in this case, then the average annual cost of running equipment will be minimized by replacing the equipment when the next periods maintenance cost becomes greater than the current average cost. Table 5.1 gives the overall maintenance cost of the refinery from 1993-2003.
CHAPTER SIX
RECOMMENDATIONS AND CONCLUSION
Recommendations
This research work has critically examined the existing maintenance procedures/culture in the new Port-Harcourt refinery as a basis for proffering solutions to the problems of low refinery products output. The work concentrated on the development and application of an optimum availability model for the new Port-Harcourt refinery, and the prescription of a replacement period for a critical component in the plant. It does not consider the numerous other possible applications of the developed model in other industries, which might bring about new specifications of the parameters. Ultimately, it has led to the prescription of a mathematical model for use in determination of the availability of the plant machines. It has equally exposed how an existing maintenance model could be applied in a new way, with the aim of obtaining better results. I therefore recommend the use of methodology of this work and its results in the new Port-Harcourt refinery, and in any other sphere whose type of business can be linked to the case study plant.
More importantly, it is recommended that each Industry in Nigeria should establish a department for data collection, analysis and visualization. This would go a long way to providing accurate data for a reliable maintenance and other related programmes policies. The department when established will serve as a direct link between research institutes/universities and the industry. The current trend of negligence of data in most of our industries drastically delay the pace of research work anchored in them (field data) and is capable of discrediting their final results.
Conclusion
In this work, effort was made to use the result of the availability analysis and other existing models in the literature to prescribe a replacement period for the critical equipment whose failure and resulting downtime override that of other units in the case study plant. Among other merits, the work has led to the presentation of a generalized availability model, maintenance management, as well as the prescription of an optimum replacement time for the turbo-generators in the Port-Harcourt refinery. The work, though generalized, made use of operational and interview data sourced from the Port-Harcourt refinery, Eleme, Rivers State, Nigeria to obtain the maintenance policies prescribed. Even though the original design was only applied to the new Port-Harcourt refinery, it can easily be tailored to suit other spheres with similar parameters.
REFERENCES
- Agunwamba, J. C. (1995), Effective Maintenance of Rural Boreholes. Water Resources vol. 6,. 1 and 2, 7-14.
- Barlow, R.E Hunter, L.C Proschan,F (1963).Optimum Checking Procedures.Siam Journal. Volume 11, 1079-1095.
- Barnes, W.J.and Brennan, J.J. (1976), An improved Algorithm for scheduling Jobs on identical machines. AIIE Transactions, 9, No 1, 25-30.
- Beebe,R.S.(2004), Predictive Maintenance of Pumps using condition Monitoring. Elsevier Advanced Technology. The Boulevard, Lang Ford Lane, Kinington, Oxford OX5 IGB,UK. P6.
- Christer,A.H,and Scarf,P.A. (1994), A Robust Replacement Model With Applications To Medical Equipment. Journal of Operation Research Society, 45.261-275.
- Christer. A.H,and Goodbody, W.(1980), Equipment Replacement in an unsteady Economy,j0urnal of operation Research Society,31.497-506.
- Christer.A.H, and Doherty,D.(1977), Scheduling Overhauls of Soaking Pits, Operational Research Quarterly, Vol.28, pages 1970 to 1977.
- Ebeling,C.E. (1997), An introduction to Reliability and Maintenability Engineering. McGraw-Hill book Company-Singapore.
- Elmaghraby,S. H,Park, (1974), Scheduling Jobs on a number of identical Machines, AIIE Transactions, V6 isue 1.
- Eti,M.C.and NWOSU,H.U,(2006), Optimum Preventive Maintenance Replacement Time for Afam Thermal Power Station. Journal of Engineering and Applied Sciences, 25, No.1, 57-65.
- Eti, M.C.and NWOSU,, H.U,(2005), Optimum Replacement Time for Pumps in Nigeria Electric Power Station. Nigeria Academic forum v.2, No.2
- Federgruen, A and SO, K.C.(1989), Optimal Policy in Maintenance cost problems. Operational Research Quarterly, 15, 813-819.
- Federgruen, A. and So, K.C. (1990), Optimal Time to Repair a broken Server Adv.Appl.Prob.21, 376-397
- Karassic, I.J, Mcguire, T. (1998), Centrifugal Pumps. Second edition, Chapman& Hall, USA.
