Pessimistic cost-sensitive active learning of decision trees for profit maximizing targeting campaigns

In business applications such as direct marketing, decision-makers are required to choose the action which best maximizes a utility function. Cost-sensitive learning methods can help them achieve this goal. In this paper, we introduce Pessimistic Active Learning (PAL). PAL employs a novel pessimistic measure, which relies on confidence intervals and is used to balance the exploration/exploitation trade-off. In order to acquire an initial sample of labeled data, PAL applies orthogonal arrays of fractional factorial design. PAL was tested on ten datasets using a decision tree inducer. A comparison of these results to those of other methods indicates PAL’s superiority.     

A paired comparison experiment for gathering expert judgment for an aircraft wiring risk assessment

Wire failure in aircraft can be attributed to several factors and the assessment of the risk of wire failure is becoming an increasingly important task. This paper will discuss the results of an actual experiment to use the paired-comparison technique for expert judgment to develop a relationship for the probability of wire failure as a function of influencing factors in an aircraft environment. The reasons for using this technique are two-fold. First, the failure probability depends on many variables including wire gauge, vibration, environmental condition, etc. In addition, the wire failure data are sparse and fitting these data to a complex failure function is a nontrivial task that may involve a host of assumptions that may not be provable.We describe a method for using actual failure data and the results from a paired comparison to populate the model parameters. In the approach, paired comparison data from select environments is used to obtain failure rate estimates for the candidate environments. Next, a functional relationship for wire failure as a function of the environments is constructed using a proportional hazards model. A regression model is fit from the failure rate estimates to the environmental variables and is used as an estimate of the failure response surface. This technique is being investigated as a means to generate failure rates for an electrical wiring interconnection system (EWIS) risk assessment software tool currently being developed for the FAA Tech Center.     

Benchmarking Iterative Receivers for BICM Multi-Carrier Systems Against Capacity Bounds

In this paper, we analyze iterative receivers for bit-interleaved coded modulation (BICM) multi-carrier systems and compare them against theoretical capacity bounds for the channel, coded modulation, and BICM. We map the theoretical capacity bounds into bit-error rate (BER) versus average signal-to-noise ratio per bit plots to simplify the comparison between the theoretical capacity bounds and simulated BER curves. As BER simulations show, iterative receivers with code doping or spreading reach the turbo-cliff within 1 or 0.3dB of the independent Rayleigh fading channel capacity. While the iterative receiver with spreading is closer to the channel capacity than the one with code doping, the later one can eliminate the residual bit-errors after the turbo-cliff. We further present a combinatorial analysis of the distribution of the spread symbol constellation for Walsh-Hadamard spreading codes used in a BICM multi-carrier system to explain the above results.     

Impact of synchrotron radiation on fuel cell operation in imaging experiments

For the clarification of water transport mechanisms in operating fuel cells, synchrotron radiation computed tomography (SR-CT) was applied. A novel fuel cell has been designed that exposes the entire active area (7 mm²) to the SR beam while at the same time allowing for full fuel cell operation during the imaging experiment. This micro fuel cell has been qualified successfully prior to the SR imaging experiments. The cell voltage was 600 mV at 0.2 A cm⁻² (open circuit voltage, OCV > 950 mV) and the operation was stable for hours. However, under SR beam exposition for in situ imaging, severe cell performance degradation within minutes has been reproducibly observed. Even after the SR beam had been switched-off cell operation remained irreversibly degraded whilst OCV could be recovered. Preliminary results indicate a higher degradation sensitivity of the cathode side of the cell. Apparently, the unique setup of the experiment which exposes an operating fuel cell with the entire active area to the SR beam reveals otherwise unnoticeable degradation mechanisms. It may have to be concluded that the very same materials degrade heavily during beam exposure that are subject of the imaging investigation. Consequently, the applicability of SR imaging to study water transport in porous fuel cell materials has to be revisited critically. This publication describes the observations made during fuel cell operation under SR beam exposure and discuss potential mechanisms that may cause beam-induced performance degradation.     

Enhanced Sliding Wear Resistance of Technical Alloys by Hard Particle Reinforcement Using Machine Hammer Peening

Machine hammer peening is a surface treatment technique originally developed for smoothening tools and mold surfaces. Treated surfaces are locally cold-worked, which results in a hardness increase and the induction of compressive residual stresses. In the present work, the feasibility of using this technique as a tool for embedding tungsten carbide hard particles on engineering-relevant substrate materials is systematically investigated. Tungsten carbide particles of three different sizes were embedded onto selected substrates using machine hammer peening. The particle embedment quality of the engineered surfaces was evaluated and correlated to the substrates' mechanical properties. The resulting tribological performance was investigated under reciprocating sliding conditions and the dominant wear mechanisms were correlated with the diameter of the embedded particles. The results show that machine hammer peening is a suitable technique for embedding hard particles in substrates of various materials, which additionally results in an enhancement in wear resistance, thus opening up a wide range of potential applications in tribologically loaded surfaces.