Statistical Approach In Order Research Articles

A statistical human resources costing and accounting model for analysing the economic effects of an intervention at a workplace

The study had two primary aims. The first aim was to combine a human resources costing and accounting approach (HRCA) with a quantitative statistical approach in order to get an integrated model. The second aim was to apply this integrated model in a quasi-experimental study in order to investigate whether preventive intervention affected sickness absence costs at the company level. The intervention studied contained occupational organizational measures, competence development, physical and psychosocial working environmental measures and individual and rehabilitation measures on both an individual and a group basis. The study is a quasi-experimental design with a non-randomized control group. Both groups involved cleaning jobs at predominantly female workplaces. The study plan involved carrying out before and after studies on both groups. The study included only those who were at the same workplace during the whole of the study period. In the HRCA model used here, the cost of sickness absence is the net difference between the costs, in the form of the value of the loss of production and the administrative cost, and the benefits in the form of lower labour costs. According to the HRCA model, the intervention used counteracted a rise in sickness absence costs at the company level, giving an average net effect of 266.5 Euros per person (full-time working) during an 8-month period. Using an analogue statistical analysis on the whole of the material, the contribution of the intervention counteracted a rise in sickness absence costs at the company level giving an average net effect of 283.2 Euros. Using a statistical method it was possible to study the regression coefficients in sub-groups and calculate the p-values for these coefficients; in the younger group the intervention gave a calculated net contribution of 605.6 Euros with a p-value of 0.073, while the intervention net contribution in the older group had a very high p-value. Using the statistical model it was also possible to study contributions of other variables and interactions. This study established that the HRCA model and the integrated model produced approximately the same monetary outcomes. The integrated model, however, allowed a deeper understanding of the various possible relationships and quantified the results with confidence intervals.

Demand side management for water heating installations in South African commercial buildings

The largest percentage of the sanitary hot water used in South African buildings is heated by means of direct electrical resistance heaters. This is one of the major contributing factors of the undesirable high morning and afternoon peaks imposed on the national electricity supply grid. Water heating therefore continues to be of concern to ESKOM, the country's only electrical utility company. The so-called in-line water heating system design methodology was developed to address this problem. This paper investigates the potential impact of in-line systems on the national peak electrical demand. A computer simulation model was developed that combines a deterministic mathematical model with a statistical approach in order to predict the diversity factors associated with both the existing and in-line design methodologies. A study was also conducted to estimate the total installed water heating capacity in the national commercial building sector. This figure can be combined with the simulated diversity factor to determine the peak electrical demand. The deterministic model includes the detailed simulation of the hot water storage vessel, the electrical heater and the system control algorithm. The mathematical model for the storage vessel is based on an electrical analogue approach that includes the effects of conduction as well as forced and natural convection. This model was verified extensively with the aid of laboratory measurements and compared with existing storage vessel models. It was found that the new storage vessel model could predict the supply temperature within 2 per cent for a system configuration with the heater in parallel outside the reservoir and within 12 per cent for a configuration with the heater situated inside the reservoir. This compares favourably with existing models found in the literature. The complete simulation based on the statistical approach showed that extensive application of the new design methodology could result in a reduction of approximately 75 MW in the total maximum peak demand imposed on the electricity supply grid in wintertime. This is 58 per cent of the current peak demand due to commercial water heating and 12.5 per cent of the peak load reduction target set by ESKOM until the year 2015. Copyright © 2001 John Wiley & Sons, Ltd.

A model for Bayesian software reliability analysis

Two problems which are of great interest in relation to software reliability are the prediction of future times to failure and the calculation of the optimal release time. An important assumption in software reliability analysis is that the reliability grows whenever bugs are found and removed. In this paper we present a model for software reliability analysis using the Bayesian statistical approach in order to incorporate in the analysis prior assumptions such as the (decreasing) ordering in the assumed constant failure rates of prescribed intervals. We use as prior model the product of gamma functions for each pair of subsequent interval constant failure rates, considering as the location parameter of the first interval the failure rate of the following interval. In this way we include the failure rate ordering information. Using this approach sequentially, we predict the time to failure for the next failure using the previous information obtained. Using also the relevant predictive distributions obtained, we calculate the optimal release time for two different requirements of interest: (a) the probability of an in-service failure in a prescribed time t; (b) the cost associated with a single or more failures in a prescribed time t. Finally a numerical example is presented. Copyright © 2000 John Wiley & Sons, Ltd.

The Problem of Regionalization in the Study of Magnitude-intensity Correspondences in the Mediterranean Area: A Distribution-free Approach

—A distribution-free statistical approach is proposed for tackling the problem of regionalization devoted to the study of magnitude-intensity relationships in the Mediterranean area. The training-set constitutes a compilation of more than 2000 earthquakes which occurred in the Mediterranean region since the end of the nineteenth century, when instrumental data became available, with epicentral or maximum intensity ≥ VI and MS magnitude values. Firstly an empirical magnitude-intensity correspondence has been computed for each intensity class by using the entire data set in the assumption of homogeneity at the regional scale. Residuals of such relation have been analyzed by a distribution-free statistical approach in order to evaluate the opportunity of a regionalization able to locally improve the performances of magnitude-intensity relationships. The analysis indicates that data concerning larger earthquakes (intensity ≥ VII) do not suggest the opportunity of zonation, and that unbiased estimates of macroseismic magnitude can also be obtained in the assumption that magnitude-intensity correspondences are uniform over the entire Mediterranean area. Therefore, better constrained relations determined for the entire Mediterranean region should be preferred to ill-defined local ones. As concerns smaller events (intensity VI), the procedure suggests that medium/small-scale lateral variations (on a wavelength lower than 102 km) should be taken into account if an efficient estimate of magnitudes from maximum observed intensity is pursued, but that data presently available are not sufficient to suggest any reliable zonation of the area under study.