Recently, the term “Cost of Ownership” has been appearing in various papers about a number of topics, including wind power and electric vehicles among others. The Wall Street Journal had an article comparing the emissions cost of a gasoline car and an electric one [Ref. 1]. But, what does the Cost of Ownership or COO really mean.
In the late 1970s and the early 1980s, the basic concept of COO was developed to provide a comparison of the generation of electricity by coal versus by nuclear. As witnessed today in China, coal fired generation plants can be constructed for a few millions of dollars in a short period of a year or less. Nuclear, on the other hand, takes a significant investment of multiple billions of dollars and a planning and construction process that lasts many years. If only construction costs are considered, the coal fired approach is much better. (Environmental impact was not as strongly considered as it is today.)
This concept – COO – was further developed by the semiconductor industry in the 1980s to justify the change from a “silk-screening” type of creating the small patterns for the circuitry to a projection of light through a mask to create the same patterns. Each wafer (the material used on which the circuits are created) contained multiple individual complete circuits that are separated into individual devices. The screening process involved tools that cost $10,000 to $20,000. The optical projection tools were in excess of $100,000. There was a throughput advantage of the optical system but not by the factor of 5 or more differential in initial price.
An evaluation of the actual cost of the two different processes was developed. The process for creating the actual circuitry involves multiple steps of imaging and processing to develop the layers of material that create the circuit functionality. A difference in yield can be quantified and that is part of the cost of ownership. The costs include the lifetime of the equipment, the materials actually consumed, operating costs, and the actual yield of the product produced.
As an example, assume the optical method produced one more good wafer equivalent per hour and each wafer was worth $1,000 with each good device worth $50. (The optical process created more good devices on an equivalent number of wafers compared to the screening process.) If the system ran only 240 hours per month, the extra value of the optical process would be $240,000 per month! That makes the $100,000 cost of the optical tool was not an issue. Consequently, the semiconductor industry changed its process.
In the 1990s and 2000s, there was further refinement in the semiconductor industry to incorporate more variables including repair costs and the resultant value of the loss of product. This COO process is moving to a total life cycle cost of ownership evaluation. As the approach is applied to new industries, the inclusion of the initial costs to create the tool/process and the dismantling impact at the end of life for the equipment. This end-of-life costs are being raised by various people regarding the batteries in general. The standard lead-acid battery carries a replacement charge when a new battery is installed. This is to cover the cost of separating out the component materials. Since there is a finite life of all batteries, one would assume that the batteries projected for storing electricity during non-peak hours will have to be replaced also.
It is not unreasonable to see further developments in costing analysis to start to include the impact of the elimination of various technology products. This might include the cost of jobs lost. It would be difficult due to the non-measurable cost of a job and whether it could be replaced by a different set of skills. COO is a good tool for evaluating alternatives but needs to be applied consistently and with assumptions that can be measured.
References:
