March 27, 2011
Update on Japan
A few quick links:
- Toll of dead and missing is in the tens of thousands, with hundreds of thousands still in temporary shelters.
- Dangerous reactor situation persists as evacuation area broadened.
- Automakers face supply problems.
- Impressive photos of rapid road repair.
- Hideaki Akaiwa: a hero's epic told in a modern style.
Posted by James Hamilton at March 27, 2011 04:30 AMdigg this | reddit
It's good to see some countries know the priority roads have in sustaining a civilization.
Posted by: aaron at March 27, 2011 08:26 AM
Globalisation and lean just-in-time production can not withstand external shocks. There has to be more slack in the system. It has become too effective.
The supply chain disruptions multiply via bull-whip effect as capacity utilization in some channels is close to 100%.
The same is true for over-leveraged global other industries and countries. They can not handle external shocks. Which are coming one after another . Very poor risk management, based on Gaussian assumptions about the probability of large oulier events. World does not follow Gaussian distribution almost nowhere. Not in Earthquakes nor in financial markets nor in supply chain bottlenecks in lean supply chain management with minimal overall stocks the main parameter.
Time to add some waste into all systems, go for less risky efficiency.
Posted by: Ivars at March 27, 2011 09:19 AM
Ivars, the evidence contradicts your thesis. Compare Haiti, China, and Japan.
Posted by: aaron at March 27, 2011 01:14 PM
If Japan kept large stocks, wouldn't we all be in worse shape now?
Posted by: aaron at March 27, 2011 01:16 PM
No, in better shape as supply chains would be still working while those dependable on them relocate them/find alternative suppliers.
Even in Fukushima, there was no slack ( redundant safety) for the case of non-Gaussian event. Supposed to happen once in million years, happened 40 years after plant design. Not clever to be so efficient, in the long term.
What has Haiti to do with some extra slack in the system? They just do not have the system, only slack. That is not what I meant.
Too effective, too overleveraged, too eager to squeeze maximum out of every penny, short term. 30-40 years is short term.
Posted by: Ivars at March 27, 2011 02:32 PM
The Fukushima incident has exposed a major case of regulatory capture in Japan. TEPCO supplies power to a third of Japan and is responsible for all of the power in Tokyo. Over the past two weeks we have seen this situation turn out consistently worse than TEPCO and the Japanese government authorities have predicted. We have also seen the US NRA openly contradict the Japanese nuclear regulatory agency and WHO openly contradict the Japanese health authorities. And Japanese officials still are only ranking this disaster a 5 on the 7 point INES scale, the same level as Three Mile Island. (Give me a break!)
Based on data collection by the Austrian Central Institute of Meteorology and Geodynamics (ZAMG) using the worldwide network of radiation detectors designed to spot clandestine nuclear bomb tests in order to monitor the compliance with the Comprehensive Test Ban Treaty (CTBT) it is now safe to conclude that in terms of the total amounts of iodine-131 and caesium-137 emitted this is already the worst nuclear energy disaster in history. As a consequence of that, and data collected by France’s radiation protection agency (IRSN), Dr Helmut Hirsch, in analysis prepared for Greenpeace, strongly recommends classifying this disaster as a 7 on the INES scale (in fact it is equivalent to three level 7 INES incidents):
The emissions could go on for weeks or even months, and given that there is roughly six times as much fissionable material at Fukushima as at Chernobyl, and that the population density of Japan is over 20 times that of Belarus, this disaster has the potential to make Chernobyl look like child’s play in terms of the human and economic toll.
The following blog post, if anything, gravely understates the potential severity of the situation:
Posted by: Mark A. Sadowski at March 27, 2011 02:38 PM
Ivars One of the problems with contemporary supply chain management is that it is too often driven by the capabilities of information management systems and the naive intuitions of IT guys rather than according to the dictates of sound inventory theory. You will find very few (if any) commercial ERP systems that have robust multi-item multi-echelon (MIME) optimizing capabilities. And what few pretenders that are available are so theoretically naive as to be useless. The usual practice is to assume a normal leadtime distribution and set a common fill rate target across all items. Of course, this is nuts. Most inventory risks typically follow a negative binomial distribution, but more importantly you want to trade-off supply performance across stocked items of supply and across supply echelons. If you're lucky you might find an ERP system that uses Lagrangian techniques to optimize across one echelon, but optimizing up through the supply chain is almost unheard of in the commercial world.
Posted by: 2slugbaits at March 27, 2011 03:08 PM
Supplies would be damaged, inaccessible. Food would be contaminated or infected.
Posted by: aaron at March 27, 2011 04:12 PM
The amount of radiation isn't as important as the types of molecules. Iodine stays in the thyroid for an extended time and cesium ends up in bones for years.
Hydrogen breaks-down fast, it stays in the body about a week. You pretty much only have a problem if you get overdosed or are unlucky enough to get cancer during that week (assuming non-repeat exposures).
Posted by: aaron at March 28, 2011 05:33 AM
"Toll of dead and missing is in the tens of thousands"
Somebody catch me up, how many of those deaths were radiation related?
Posted by: KevinM at March 29, 2011 07:49 AM
Kevin, to the best of our knowledge as of 3/29/11, the answer is zero.
Posted by: HB at March 29, 2011 12:26 PM
apanese domestic car sales were down about 30% in March from year-earlier levels, according to reported preliminary data released Wednesday by the Japan Automobile Dealers Association. The figures covered sales for the first 29 days of the month, and were reported by Nikkei. Only 230,000 new cars were sold in the period, the report cited the association data as saying. Little change was expected in full-month figures, as a surge in auto sales normally seen in the final days of the month is unlikely owing to vehicle-inventory shortages arising from quake-related factory closures. Seperately, Credit Suisse said Wednesday that Japanese industrial production is set to decline 10% in March from February's figures. The prefectures in northern Japan most affected by the twin disasters account for 7.3% of gross domestic product, but the related supply-chain disruptions are resulting in shortages throughout Japan's industrial base and magnifying the slowdown, Credit Suisse said.
Posted by: ivars at March 30, 2011 12:01 AM