On the night of 17th Dec 2011 (Sat), engineers from SMRT and the LTA performed a thorough examination the MRT system and found 21 damaged claws which secure the third rail, a rail that provides traction power to the train. So ostensibly, the technical cause lies in the power transmission system.
However, the questions should not stop there, not until it is understood what would have allowed this technical fault to cause so many severe operational failures within such a short time. The apparent point of failure, the third rail, is , I am told, a very sturdy structure.
Furthermore, noting that the tracks “have a lifespan of 50 years” and that “there is no recommended time frame on how often these tracks have to be checked”, it is clear that we are dealing with equipment with long inter-failure intervals.
Long intervals between failures can give a false sense of security. The illusion of 10 to 20 years of smooth operation might be extrapolated in an unwarranted fashion. The obvious corollary to this logical fallacy is the reduction in frequency of preventive maintenance because (i) it appears to add little value to the company, and (ii) it costs money.
Let us now consider the environment in the forum where decisions on maintenance policy are ultimately made. The boardroom is a very quantitatively rational place. It is not far from the truth to state that most business decisions are made by reducing expected consequences of each candidate decision to a number and choosing a course of action based on which decision is associated with the largest number.
Senior management in public companies are under heavy pressure from the bottom-line, as their Profit & Loss statement has a major impact on their careers and their remuneration. This is why decisions such as retrenchment and off-shoring are made even when survival is not at stake. Even though business students are universally warned to guard against “short-term pressures”, most eventually succumb when it is their turn to step up to the plate. It is hard to be self-righteous about “acting in the long-term interests of the company” when one is aware of the real pressures that decision makers face in a public company.
When Senior Management is to make a decision about maintenance policy, the following information might be presented to them: (i) the suppliers’ recommended preventive maintenance frequency, (ii) local historical information about failures and their financial consequences, (iii) a set of alternative policy parameters, and (iv) a risk assessment for each alternative. Where risk assessments for recommended policy parameters seem too conservative when contrasted with historical information, riskier choices tend to be made. It is speculated that this has been the case for SMRT.
At this point, the shareholders of SMRT, in particular the Singapore Government as majority stakeholder, should decide on what constitutes good performance of the MRT system and take steps to align the behavior of SMRT senior management in that direction. If it means increasing system availability by inducing more aversion to the risk of operational failures, then so be it. To do this, suitable KPIs have to be measured and a sufficiently large fraction of senior management remuneration should be linked to good performance on those KPIs.
The only form of trickle-down economics that has been historically shown to work turns out to be very Confucian in structure — by rectifying the incentives of senior management, eventually performance measurement for middle management will become more aligned to corporate goals, resulting in line operations being carried out in a manner that serves these goals, leading to better overall systemic performance. This might be dubbed “the Great Change Management”, and is recommended as the way forward for SMRT.
Presently, SMRT has identified a stop gap measure: To run trains more slowly over certain stretches. Hopefully this gives the LTA and SMRT time to work things out top to bottom, from the boardroom all the way to the power rail.