Potential for Improvements—Productivity

Issue of permits to work (PTW) can be a major problem affecting productivity in any plant. In this shutdown, operations had streamlined their work very well. Between 3 and 4 pm., they discussed the safety aspects of the following day's work with the relevant contract supervisors. The PTWs were ready in all respects except for the signatures of the two parties. Next morning, operations merely updated the contract supervisors with any changes in safety requirements and then signed it. They asked them to meet the field operators before commencing work. Meanwhile, they contacted the field operators so that they could ensure that the work commenced safely. Using this procedure, they were able to release, on average, three PTWs per minute.

On day 4 of the shutdown, an audit showed that by 8 am., only ten permits had been released. The distribution can be seen in Figure 36.1. In a discussion with the shutdown leader, a solution that emerged was to ask the contract supervisors to come in earlier, at 7.30 am, allowing them to have the permits in hand before their crew arrived. This plan was implemented and an audit on day 8 showed a slight improvement. An audit on day 15 showed that the problem had been substantially resolved. These results can be seen in Figures 36.2 and 36.3.

Figure 36.1 Issue of PTWs on Day 4

Issue of PTWs on Day 8

Figure 36.2 Issue of PTWs on Day 8

Issue of PTWs on Day 15

Figure 36.3 Issue of PTWs on Day 15

Another significant cause of loss of hand-on-tools time was the way working hours are organized. At this location, shutdown work for non-critical activities was done on an eight-hour day-shift basis. Contractors transported their workers from local bus stations, designated car parking lots, and the refinery gate to the central washroom area. Then the workers walked to the shutdown area. The process was repeated in reverse at the end of the day. Each of these walks took about 7-8 minutes.

Supervisors gave tool-box talks to their crew before they left for their work site. These talks were quite short, lasting 2-3 minutes. Since each supervisor had 3-5 crews, this process could take 15 minutes. It took 3-5 minutes to walk to the work site, sometimes longer if they needed to climb stairs or scaffolding. By the time the crews started work, it was invariably about 8:45 am.

Contractors were responsible for providing coffee and tea to their workers. They used vans for this purpose. The first of these arrived at 9:30 in the morning and 2:15 in the afternoon. Human nature being what it is, as soon as the first van was sighted, someone whistled, and everybody came down. This meant that the first work period could not exceed 45 minutes. After the morning break, the crews could work uninterrupted till 11:45. They were allowed 15 minutes to wash up before lunch at noon. The afternoon coffee break lasted 30 minutes instead of 15, for the same reasons as in the morning. Traditionally, the workers had been allowed 45 minutes at the end of the day, to allow time to wash up and walk to the gate. The central washroom was crowded and buses left promptly, so extra time was needed to ensure that the workers reached in time. This meant that after the afternoon coffee break they could work only 11/2 hours. Thus, the available time to work was only 51/2 hours. Of this, they worked perhaps for 31/2 hours. This type of situation must have been experienced by a number of readers in their own work sites.

The main players in the shutdown team discussed the results of the analysis, and evaluated possible solutions. I suggested we add 11/2 hours of planned overtime every day, making it a 91/2 hour day. They rejected it immediately. In continental Europe, there is strong social legislation. Overtime work needs governmental approval, and planned overtime needs prior approval. Team members felt that the authorities would turn down our request. Later, I discussed this with the person who dealt with the government agency concerned. He said it was perfectly feasible, exploding another myth. This was something we could explore at least for future shutdowns. The benefit of 91/2 hour vs. 8 hour days is illustrated in Figure 36.4. The main advantage is not the total additional time, but the fact that the blocks of time are longer and thus more effective.

This still left us with the problem of the 45 minutes lost at the end of the day. There was no easy solution to the washroom crowding problem. For various reasons, changing departure time of buses couldn't be discussed at that time. Traditions are hard to change, so we decided to accept this loss, at least for the present. After all, with the 91/2 hour day, we would gain 2 hours of available working time. More importantly, these were in longer and more meaningful blocks of about 11/2 hours. One could reasonably expect that the actual work output would rise correspondingly by 30-35%. This meant that the 40 days duration could in theory be 30 days. In practice, one might expect a smaller reduction, say 5 days, with a value of US$1.5m.

Benchmarking studies showed that in plants of comparable size in other countries, Operations needed much less time to shut down and start up the unit. The potential reduction was about 9 days. Environmental and safety considerations resulted in additional work, and Operations' resources were severely limited, in spite of working on a 2x12 hour basis. Another brainstorming session ensued, with some interesting ideas. In the short term, the idea

Effect of 91/2 Hour Day

Figure 36.4 Effect of 91/2 Hour Day

was to hire skilled operators from contractors during the busy shut-down and start-up periods. For the long term, an idea that was floated was to train the mechanical crew to support operations in valve operation, shutting down complex machinery, etc. After all, during the shutting-down process, mechanics were largely on standby, doing preparatory work. This idea had many pitfalls, so we parked it for later action. Contract operators would cost about US$ 200,000. A modest reduction in duration of, say, 5 days was worth US$1.5m.

Similarly, one reason for the criticality of the refractory work was because they worked 3x8 hours. Refractory masons were in short supply, so the work had to be done in sequence. Had they been able to work in parallel, this work would not be on the critical path. Two shifts of 81/2 hours each would release one complete shift crew, enabling work in parallel. Applying the same approach as earlier with the 91/2 hour day work, we could get a much higher 'available time,' thereby increasing productivity. As a result of traditional thinking, it was assumed that critical work must be done on a 24-hour basis. That this is not always true is evident from the above discussion.

During shutdowns, normal weekly working hours were 5 days x 8 hours. As a result of all these discussions, the refinery decided to work 6 days x 91/2 hours. On critical path activities they worked 2x9 hour shifts for 5 days and one 9-hour shift on Saturdays. In a few exceptional cases, they used 2x12 hour shifts. Note that skilled labor availability and prevailing legislation were limiting factors. They used the weekend to catch up on delays by doing additional overtime work.

< Prev   CONTENTS   Source   Next >