As I write this, the Australian Open Tennis tournament is being contested. A men’s singles match from the second round caught my eye. It featured Gilles Simon and Julien Benneteau, both from France. Simon won more games (25 to 24) and more points (165 to 163) but Benneteau won the match 3 sets to 2. Using different measurements, Simon would have won.
It got me thinking about measuring success in industry, and if there were “successful” results that would have been failures had other measurements been used.
When I was in operations, we had one measurement that management used to show overnight performance on the morning report – sales gas calorific value. If we made 1003 Btu/cubic foot it was considered as good as possible. If we made 1004, questions were asked. We them implemented a project involving adding a new supply field. Overtime, the plant produced 1004 Btu/cubic foot. We never could produce 1003. I stopped trying to figure out the plant, and looked at the measurement. I came to the conclusion that for this NEW feed, 1004 was optimum. I spent a great deal of effort trying to get management to accept this “new” measurement. Eventually I was successful.
For projects, the four project “pillars” of industrial success tend to be safety, quality, cost, and schedule. So is it possible to meet all measurements, but still fail? Or vice versa? I am reminded of the oil-fired electric power plant that was built as back-up to hydro-electric power in the event of prolonged drought. The power plant was built on time, under budget, and with no incidents. In addition, it passed all performance and acceptance tests. But it was a failure. It was built on the seacoast where a river entered the sea. Cooling water for the power plant was taken from the river, and returned to the sea “on the other side of an outcropping”. The river water as about 2 degrees cooler than the sea water.
Why was it a failure? It was built for periods of drought. In drought, there would be much less river flow, and the effects of the tide would raise river water temperature. So while it passed performance tests, the tests were done when the river was normal. During drought, the river flow dropped, seawater back flowed up the river, raising the temperature, and significantly reducing the capacity of the power plant. Capacity met during testing, but not when needed. Failure, with all performance measurements successfully met.
Another example (yes it is a bit of a stretch, but stay with me) is Hetch Hetchy. Hetch Hetchy is inside Yosemite National Park in California, and is sometimes considered the twin of Yosemite Valley. Hetch Hetchy was, however, “destroyed” when it was dammed in the 1920s to make a drinking water lake supply lake for the city of San Francisco. A hydroelectric generator was also incorporated into the dam.
Today Yosemite Valley is visited by over 3.7 million people each year (which must put significant environmental strain on the valley). Hetch Hetchy, on the other hand, is rarely visited. Those that have the opportunity to visit Hetch Hetchy often report seeing rare and endangered animals, with bobcat sightings being quite common.
There is currently a movement (supported by actor Harrison Ford) to drain and restore Hetch Hetchy – even though it provides San Francisco with about 25% of its drinking water supply. Restoration would also reduce electrical generation by about 280 million kWhr/yr. While some countries would welcome hydroelectric power as a carbon management tool, it looks like California will not recognise it. I have not seen the key performance indicators for this project, so I do not know if this can be a successful project or not … or if this is a project where completion at any cost equals success.