Applying genetic algorithms to dampen the impact of price fluctuations in a supply chain

Genetic Algorithms (GAs) have been identified as an innovative and useful approach for dampening the Bullwhip Effect along supply chains. This paper extends previous work by developing an improved supply chain model that incorporates additional cost factors such as ordering cost, item cost, distribution cost and production cost. The revised model is then used to examine one element of the Bullwhip Effect, i.e. price fluctuation strategies. A GA is employed to determine the ordering policy for each member in the model that minimises cost. The research illustrates how the GA performs if a sales promotion is introduced. From the experimental results, it is shown that a GA can help determine an improved ordering policy and reduce the total cost across the supply chain.     

An evaluation of the predictive accuracy of wake effects models for offshore wind farms

Wake losses are perceived as one of the largest uncertainties in energy production estimates (EPEs) for new offshore wind projects. In recent years, significant effort has been invested to improve the accuracy of wake models. However, it is still common for a standard wake loss uncertainty of 50% to be assumed in EPEs for new offshore wind farms. This paper presents a body of evidence to support reducing that assumed uncertainty. It benchmarks the performance of four commonly used wake models against production data from five offshore wind farms. Three levels of evidence are presented to substantiate the performance of the models:

• Case studies, i.e. efficiencies of specific turbines under specific wind conditions;
• Array efficiencies for the wind farm as a whole for relatively large bins of wind speed and direction; and
• Validation wake loss, which corresponds to the overall wake loss within the proportion of the annual energy production where validation is possible.

The most important result for predicting annual energy production is the validation wake loss. The other levels of evidence demonstrate that this result is not unduly reliant on cancellation of errors between wind speed and/or wind direction bins. All of the root‐mean‐squared errors in validation wake loss are substantially lower than the 50% uncertainty commonly assumed in EPEs; indeed, even the maximum errors are below 25%. It is therefore concluded that there is a good body of evidence to support reducing this assumed uncertainty substantially, to a proposed level of 25%. Copyright © 2015 John Wiley & Sons, Ltd.