On the effectiveness of early life cycle defect prediction with Bayesian Nets

Standard practice in building models in software engineering normally involves three steps: collecting domain knowledge (previous results, expert knowledge); building a skeleton of the model based on step 1 including as yet unknown parameters; estimating the model parameters using historical data. Our experience shows that it is extremely difficult to obtain reliable data of the required granularity, or of the required volume with which we could later generalize our conclusions. Therefore, in searching for a method for building a model we cannot consider methods requiring large volumes of data. This paper discusses an experiment to develop a causal model (Bayesian net) for predicting the number of residual defects that are likely to be found during independent testing or operational usage. The approach supports (1) and (2), does not require (3), yet still makes accurate defect predictions (an R ² of 0.93 between predicted and actual defects). Since our method does not require detailed domain knowledge it can be applied very early in the process life cycle. The model incorporates a set of quantitative and qualitative factors describing a project and its development process, which are inputs to the model. The model variables, as well as the relationships between them, were identified as part of a major collaborative project. A dataset, elicited from 31 completed software projects in the consumer electronics industry, was gathered using a questionnaire distributed to managers of recent projects. We used this dataset to validate the model by analyzing several popular evaluation measures (R ², measures based on the relative error and Pred). The validation results also confirm the need for using the qualitative factors in the model. The dataset may be of interest to other researchers evaluating models with similar aims. Based on some typical scenarios we demonstrate how the model can be used for better decision support in operational environments. We also performed sensitivity analysis in which we identified the most influential variables on the number of residual defects. This showed that the project size, scale of distributed communication and the project complexity cause the most of variation in number of defects in our model. We make both the dataset and causal model available for research use.     

Estimating thermal stress in BIPV modules

The thermal stress on building‐integrated photovoltaic modules (BIPV) in Espoo, Finland, was studied with field‐testing of amorphous silicon modules. Based on these results, the thermal stress at two other European locations (Paris and Lisbon) was estimated. The estimation procedure entailed thermal modelling of heat transfer in the façade with meteorological data as input. The results indicate that the thermal stress on BIPV modules in Lisbon is, in this case, approximately 50% higher that in Espoo and between 80 and 200% higher than in Paris, depending on the activation energy of the degradation process. The difference in stress between a BIPV module and a free‐standing module in Espoo was 50–200%. Copyright © 2006 John Wiley & Sons, Ltd.     

Design and Characterization of the CLIMPAQ,Chamber for Laboratory Investigations of Materials,Pollution and Air Quality*

A test chamber has been developed in order to provide a small and simple emission testing facility capable of testing construction products in a climate where the important climatic parameters such as temperature, ventilation rate and air velocity can be varied independently around typical indoor values. The test chamber CLIMPAQ is made of panes of window glass. Other main surface materials are stainless steel and eloxated aluminium. The chamber has a volume of 50.9 litres and is designed to meet the requirements for quantifying air pollution. In this investigation human subjects acted as air pollution judges, and chemical characterization of the air pollution was carried out. Carpet, linoleum, wall paint and seal- ant were tested simultaneously in the CLIMPAQ and in four other chambers ranging from a full-scale chamber of 28 m³ to a field and laboratory emission cell of 3.5· 10^?5m³. Product ranking is the same in all chambers for the sensory measurements. Emission rates based on sensory measurements differ for all products less than 100 % except for tests in a 3-litre chamber where emission rates were higher. Chemical measurements differ up to approximately 10 times for the same product in different chambers. Deviations appear to be the result of different environmental parameters in the various chambers. Low air concentrations or high specific ventilation rates seem to increase emissions, while differences in air velocities and sink properties may also be the cause of differences in emission rates.     

Personal Exposure in Displacement Ventilated Rooms

Personal exposure in a displacement ventilated room is examined. The stratified flow and the considerable concentration gradients necessitate an improvement of the widely used fully mixing compartmental approach. The exposure of a seated and a standing person in proportion to the stratification height is examined by means of full-scale measurements. A breathing thermal manikin is used to simulate a person. It is found that the flow in the boundary layer around a person is able to a great extent to entrain and transport air from below the breathing zone. In the case of non-passive, heated contaminant sources, this entrainment improves the indoor air quality. Measurements of exposure due to a passive contaminant source show a significant dependence on the flow field as well as on the contaminant source location. Poor system performance is found in the case of a passive contaminant released in the lower part of the room close to the occupant. A personal exposure model for displacement ventilated rooms is proposed. The model takes the influence of gradients and the human thermal boundary layer into account. Two new quantities describing the interaction between a person and the ventilation are defined.     

German Materials Science: A self-portrait

An influential and refreshingly frank report on the state of health of non-university materials research in Germany has recently been published. National Research Centers, Max Planck Institutes, and Fraunhofer Institutes are among the organizations covered. Recommendations are made on the basis of materials type, and on the performances of the individual institutes. The conclusions of this self-portrait, painted by representatives of the organizations themselves along with participants from the universities, industry, and state, will provide food for thought for both German materials scientists and those in other European countries concerned with the future development of this crucial economic sector.     

Using two factor structures of the Mental Adjustment to Cancer (MAC) scale for assessing adaptation to breast cancer

The validity and reliability of two factor structures of the Mental Adjustment to Cancer (MAC) scale for assessing coping style was assessed by examining the relationship between the subscales, psychological distress, and quality of life (QOL) among Stage II and Stage IV breast cancer patients in four phases. First, MANOVAs assessed differences in coping, distress, and QOL across disease stages. Second, for each MAC factor structure, canonical correlation analyses assessed the relationship between coping styles, distress, and QOL, for each disease-stage group separately. Third, structural equation modeling (SEM) assessed the relationship among coping styles, distress, and QOL for all participants. Finally, the internal consistency of both MAC factor structures was assessed using Chronbach's alpha. The results were as follows: (1) significant differences across disease stages were found for coping styles using either the Watson or the Schwartz MAC subscales, but there were no differences in levels of distress or QOL; (2) for both MAC factor structures, coping style was found to be highly related to emotional distress and QOL, however, the strengths of the correlations between individual coping styles and distress/QOL indicators varied across disease stages; (3) SEM indicated that coping style was significantly related to distress and QOL when stage of disease was not considered, and that coping style and indicators of distress/QOL are separate, but highly correlated factors, as opposed to a single latent construct; and, (4) the Watson MAC subscales showed slightly better internal consistency than the Schwartz MAC subscales. Taken together, these findings highlight: (i) the validity of both MAC factor structures for clinical and research use with American breast cancer patients; (ii) the role of coping style as a mediator between disease stage and psychological distress and QOL; and, (iii) the need for refinement of certain Watson and Schwartz subscales.     

LD泵浦的声光调Q腔内倍频激光器

研制成用ＬＤ泵浦的声光调Ｑ的Ｎｄ：ＹＶＯ４－ＫＴＰ腔内倍频激光器，得到了ＴＥＭ００模、频率高达１００ｋＨｚ的稳定的５３２ｕｍ绿激光脉冲系列输出，阈值泵浦功率为２７ｍＷ；在连续５７０ｍＷ的泵浦功率下，绿激光脉冲的峰值功率达１１００Ｗ，脉宽为４．６ｕｓ；输出功率稳定，在±５℃的工作温度变化下，输出功率变化小于±２％。此种器件可望得到广泛的应用。