Computers spur maintenance-management concepts

Thi Chang
Refining/Petrochemical Editor

Reliability-centered maintenance (RCM), total productive maintenance (TPM), and risk-based inspection (RBI) were the buzz words of the National Petrochemical & Refiners Association (NPRA) Maintenance Conference May 19-22 in San Antonio.

These maintenance-management concepts are spurred by the increased speed, increased processing power, and friendly user-interfaces of personal computers. RCM, TPM, and RBI require huge data bases that were very costly to maintain 5 years ago. Many users are tying RCM to their new SAP AG systems to facilitate the accounting processes related to maintenance.

RCM

RCM is an effort focused on equipment reliability to maximize profit. It identifies the maintenance requirements for each piece of equipment to keep the operations functional. RCM teams collect historical data from every piece of equipment in the operating plant. The data analyses involve correlation of various equipment components with historical and potential failures. These data are correlated with possible preventive measures.

With wise implementation of preventive maintenance, operating companies hope to reap lower maintenance costs, reduce maintenance backlogs, and facilitate teamwork between the maintenance and the process operations in the plant.

Other names for RCM include preventive maintenance, reliability-focused maintenance, and autonomous maintenance.

Since RCM implementation in 1996, Chevron Products Co., El Paso, has achieved a reduction in work backlog, an increase in preventive maintenance, and a reduction in the number and cost of incidents (e.g., fires, spills, unplanned releases), according to Sam H. Preckett, an employee in the projects maintenance department. Although RCM is not solely responsible for these successes, according to Chevron, it played a major role.

TPM

Total productive maintenance (TPM), espoused in a paper presented by Susan E. Alvarado, TPM facilitator, and Mike A. Woolbert, quality/training manager, both from Phillips Petroleum Co., Sweeny, Tex., can help with the lack of employee buy-in.

TPM seems to be RCM with a focus on employee involvement. According to the Phillip authors, "TPM combines the American practice of preventive maintenance with total quality control and total employee involvement."

According to Doug Martin, TPM manager, DuPont Co., Camden, S.C., TPM is more than autonomous maintenance. Operators and maintenance personnel should work together to identify tasks that can be transferred. "Operators should describe tasks which they feel they could perform with their existing skills or with minor training, and the craftsmen agree that these tasks do not require the craftsman skill level."

The TPM approach encourages operators to share in the maintenance work by involving them in the planned maintenance development. Woolbert describes it as a "culture change."

"If you're writing work orders to control your people every 2 hr, you better look at your system. Beginning to have ownership is the key to success," Woolbert warns.

In their paper, Alvarado and Woolbert claim the following improvements since TPM began at the Phillips Sweeny complex in 1994: a savings of millions of dollars, improved equipment performance, a reduction of routine maintenance cost by 50%, a reduction in OSHA reportable incidents by 67%, and a reduction of inventory by 50%.

TPM has only been implemented on rotating equipment and will be expanded to fixed equipment this year. As evidence of their success, Alvarado pointed out that their last seal failure occurred 6 months ago.

The complex is currently being considered for a TPM award by the Japanese Institute of Plant Maintenance (JIPM). Phillips Sweeny complex is the first North American facility to undergo a TPM award preliminary assessment.

RBI

RBI is a quantitative method to evaluate maintenance needs. RBI is often used as a tool in maintenance decisions involving turnarounds.

The goal of RBI is to optimize the financial return of the maintenance decision. The financial return is either the benefit of doing maintenance to avoid needed repairs or the benefit of avoiding maintenance and its consequential costs.

The financial return is further expanded to account for the probability of failure of the equipment and the severity of the consequences if a failure occurs. Based on financial return, probability, and consequences, management can make a quantitative maintenance decision.

Table 1 [43,706 bytes] and Table 2 [43,688 bytes]show the results of maintenance optimization techniques performed on 11 individual plant components. This optimization analysis was presented in a paper by Daniel I. Ash, research engineer, and David A. Mauney, staff engineer, both of Southwest Research Institute in San Antonio.

Table 1 shows suggested component replacement based on a conventional engineering analysis, which resulted in a net present value (NPV) of $50,000. By combining engineering and risk (RBI), the replacement dates in Table 2 achieve an NPV of $1,214,000.

Equistar Chemicals LP is using RBI for planning an upcoming turnaround. In a careful RBI evaluation of its equipment needs, Equistar estimates a savings of about $195,000 in turnaround costs by limiting the number of internal inspections.

Lots of data

Paul Balmert, assistant plant manager at Union Carbide Corp, Texas City, Tex., urges companies to take advantage of the lowered prices of communication technology and use remote monitoring to its potential. This use, he claims, can reduce maintenance and improve reliability.

RCM collects huge amounts of data on each piece of equipment from remote monitoring. It also collects equipment repair histories, critical performance parameters, and specifications. Much of the data are used for statistical evaluations. Some of the data are not used in the analysis at all. The advances of information systems allow and encourage the collection of more data-not necessarily only the needed data.

RBI also creates a huge data base of consequences and probabilities as well as equipment inspection needs. These data are analyzed for decisions on the type and frequency of inspection that is the most cost-productive.

Leadership

A consistent recurring motif was the need for strong leadership and teamwork to make the programs work. Senior leadership support is vital for budget, morale, and consistency. If the team feels that the maintenance program is only the latest fad, the whole program suffers.

Ralph Gerstenkorn and Michael Torpey, machinery engineer and mechanical department supervisor, respectively, of Atlantic Richfield Co.'s Cherry Point refinery, Blaine, Wash., presented a paper documenting the successful results of their initial maintenance-program implementation. According to Gerstenkorn and Torpey, ARCO reduced maintenance costs of 128 steam turbines by 80% within 3 years.

They point out at the end of their presentation the need for the operators' buy-in of the maintenance initiative. Without the operators' cooperation, the task assigned to them, i.e., the lubrication of turbines to prevent expensive lubrication failures, render the maintenance goals obsolete.

"The question we are faced with is this: How do we keep a large group of operating technicians checking oil routinely?"

Caution

RCM is not always the best choice, however, claims Heinz P. Bloch, consultant, Process Machinery Consulting, Montgomery, Tex. Bloch estimates that about 60% of the U.S. industry's attempted RCM efforts have been abandoned after 1-2 years.

He advocates using RCM only after the basic reliability improvement techniques have been implemented. Troubleshooters should be properly trained and well-read, not necessarily have RCM knowledge.

Bloch advocates back to the basics: "Only equipment that exhibit[s] wear-out failures will optimally respond to RCM...random failures are best detected by predictive maintenance, experience, and instrument-based gathering and analysis techniques."

Bloch points out the ineffectiveness of RCM in an example. With the knowledge that centrifugal-pump bearings fail prematurely as a result of atmospheric dust and moisture, the maintenance solution is to hermetically seal the bearing housings with a magnetic face seal. RCM and other resource-consuming techniques are not necessary to economically solve this problem.

In a paper presented by Laurence M. Myers, RWD Technologies Inc., Houston, Myers points out several reasons RCM deployment has failed. One of those reasons is failure to involve the end users, who are the operations and maintenance personnel.

"The key to success is that many personnel from all levels of the organization must be involved to ultimately create a program that yields greater alignment with corporate goals."

Similarly, Martin warns that failure of TPM is certain, as a result of distrust, if the goal is the reduction of craftmen jobs and not the pursuit of autonomous maintenance.