A new approach: Interrelated two-way clustering of gene expression data

The paper presents a new approach to interrelated two-way clustering of gene expression data. Clustering of genes has been effected using entropy and a correlation measure, whereas the samples have been clustered using the fuzzy C-means. The efficiency of this approach has been tested on two well known data sets: the colon cancer data set and the leukemia data set. Using this approach, we were able to identify the important co-regulated genes and group the samples efficiently at the same time.     

Construction productivity

Constraints impeding the growth of construction productivity, originating largely from the fragmented structure of the construction industry are reviewed. Technological advances in design and construction have always contributed to significant increases in productivity at all levels. However, recent trends with regard to total industrial productivity (TIP) are disturbing and may reflect the dominant but adverse influence of external economic factors and their interference with the construction process; there seems a definite connection between low productivity and low industrial output. There is always a potential conflict between enhancing TIP and efforts to improve productivity in the various segments of the industry. Stagnation or further decline in TIP will lead to still higher construction costs and reduced demand on the industry in the future. The need is for an integrated management approach to strengthen the total industrial framework, especially to resist the unfavourable effects of fragmentation and unpredictable economic conditions. An example of such an approach from Japan's sector is described. Managerial efforts aiming solely at greater efficiency on job sites or in firms will produce only marginal benefits, especially if contemporary economic conditions continue.     

A sample size selection procedure for pairwise comparisons of means with unequal variances

A sample size selection procedure for paired comparisons of means is presented which controls the half width of the confidence intervals while allowing for unequal variances of treatment means.     

Cost Efficiency Benchmarking for Operational Units with Multiple Cost Drivers

We consider the activity‐based costing situation, in which for each of several comparable operational units, multiple cost drivers generate a single cost pool. Our study focuses on published data from a set of property tax collection offices, called rates departments, for the London metropolitan area. We define what may be called benchmark or most efficient costs per unit of driver. A principle of maximum performance efficiency is proposed, and an approach to estimating the benchmark unit costs is derived from this principle. A validation approach for this estimation method is developed in terms of what we call normal‐like‐or‐better performance effectiveness. Application to longitudinal data on a single unit is briefly discussed. We also consider some implications for the more routine case when costs are disaggregated to subpools associated with individual cost drivers.