Logistic model-based forecast of sales and generation of obsolete computers in the US

Our goal is to characterize future trends in the generation of obsolete computers in the U.S. Starting from historical sales data on new computers and assuming a plausible first lifespan distribution, we extrapolate the historical sales trend to the future using a logistic model. The major challenge is that the personal computer is still in an early stage of its technology adoption life cycle and thus early for statistical fits to yield a reasonable estimation of carrying capacity (or saturation adoption level). Our approach is to use a bounding analysis which characterizes a range based on plausible upper and lower bounds on the future carrying capacity (1.3 and 1.0 computers per capita respectively). These lower and upper bounds yield a generation of 92 and 107 million obsolete computers in 2020 respectively. The growth rates of adoption over the next decade are very different for lower versus upper bound, however by 2020 the adoption will be at most 8% away from the long-term carrying capacity in both cases. Assuming computer adoption follows logistic behavior we assert that the saturation level of generation of obsolete computers is not much more than a decade away. The current recycling level of computers is 65 million units, thus if the U.S. expects to recycle computers domestically significant growth of recycling facilities will be required. Note however that this analysis does not address how long obsolete computers are stored nor the distribution of obsolete computers to reuse, recycling, landfill options. This is an important issue to resolve in future work. (C) 2009 Elsevier Inc. All rights reserved.