Not only is the cost of conventional solar modules declining, as shown in Figure 6.5, but a new solar technology, known as concentrating photovoltaics, or CPV, is emerging as a potentially lower cost competitor. The data file ch06_MCSP_Solar gives data on the cost of CPV in dollars per watt of installed capacity for the 19 major installations that took place during 2007–2013, from a study by the SunLab at the University of Ottawa. It is clear that the cost came down rapidly during this early period in the commercialization of CPV. But why do costs come down over time? Is it just the passing of time itself, or is it due to the industry gaining experience with the new technology and figuring out ways to implement it more efficiently? If it is the latter, how can we measure “experience”? We need to understand what it is that leads to a decline in costs so that we can project future costs and plan when to install a CPV project. The data file also gives information on the volume of CPV that has been installed to date (in megawatts, MW) as a measure of how much experience the industry has accumulated as each major installation is completed.
Plot (i) cost against time and (ii) cost against cumulative volume, and describe the scatterplots you obtain. Can a correlation coefficient be calculated for these data? Take the logarithm of cost and answer the same questions. Finally, plot the logarithm of cost against the logarithm of cumulative volume and answer the same questions. This final log/log plot is known as the “experience curve” and has been used to track how cost depends on experience for a wide range of technologies, including microwave ovens, light bulbs, and military equipment. Which of your graphs would you choose as the best means of showing the correlation in the data about CPV? Give your reasons.