Abstract
The web benchmarking systems broadly used in the construction industry (CI) are designed to provide results based on key performance indicators (KPIs). No insights concerning organization overall performance and improvements targets are available. This research aims to fulfill this gap using data envelopment analysis (DEA) as a method to complement the information provided by a set of KPIs. The methodology proposed is useful to all organizations involved in benchmarking routines. To enable a more realistic assessment of CI companies, two types of DEA models were used, one allows factor weights to vary freely and the other includes weight restrictions. These models assign an efficiency score to each organization, identifying efficient organizations and providing performance improvements targets for the others. To enable suggesting targets for all organizations, expert opinion was used to specify virtual units which were included in the efficiency assessment to define a practical frontier located beyond the productivity levels of the original DEA frontier. Based on a sample of 20 Portuguese leading contractors, the Portuguese web benchmarking system for CI, icBench, was used to demonstrate the advantages of integrating the DEA method with KPIs benchmark scores.

Abstract
This paper analyses the value added (VA) of a sample of Portuguese schools using two methodologies: data envelopment analysis (DEA) and the methodology used presently by the UK Department for Children, Schools and Families (DCSF). The VA estimates obtained by the two methods are substantially different. This reflects their different focus: DEA emphasizes on best-observed performance, whereas the DCSF method reveals average performance. The main advantage of the methodology used by the DCSF is its simplicity, although it confounds pupil effects with school effects in the estimation of school VA. In contrast, the DEA methodology can differentiate these effects, but the complexity may prevent its use in a systematic way. This paper shows that the two methods provide complementary information regarding the VA of schools, and their joint use can improve the understanding of the relative effectiveness of schools regarding the progress that pupils make between educational stages.

Abstract
Recently Tracy and Chen presented a parametric DEA model (PDEA) to assess relative efficiency in the presence of a generalized form of linear weight restrictions. This paper proposes a modification to the PDEA model that avoids the need to resort to searching algorithms to estimate efficiency, and assures that the correct efficiency scores are obtained in a single stage using mathematical programming solvers. The results of this model and the results of Tracy and Chen's PDEA model are compared using the examples reported in their paper. The results confirm the superiority of the model proposed in this paper. Journal of the Operational Research Society (2010) 61, 1789-1793. doi:10.1057/jors.2009.140 Published online 16 December 2009

Abstract
The most popular weight restrictions are assurance regions (ARs), which impose ratios between weights to be within certain ranges. ARs can be categorized into two types: ARs type I (ARI) and ARs type II (ARII). ARI specify bounds on ratios between input weights or between output weights, whilst ARII specify bounds on ratios that link input to output weights. DEA models with ARI successfully maximize relative efficiency, but in the presence of ARII the DEA models may under-estimate relative efficiency or may become infeasible. In this paper we discuss the problems that can occur in the presence of ARII and propose a new nonlinear model that overcomes the limitations discussed. Also, the dual model is described, which enables the assessment of relative efficiency when trade-offs between inputs and outputs are specified. The application of the model developed is illustrated in the efficiency assessment of Portuguese. secondary schools.

Abstract
The problem of predicting the next request during a user's navigation session has been extensively studied. In this context, higher-order Markov models have been widely used to model navigation sessions and to predict the next navigation step, while prediction accuracy has been mainly evaluated with the hit and miss score. We claim that this score, although useful, is not sufficient for evaluating next link prediction models with the aim of finding a sufficient order of the model, the size of a recommendation set, and assessing the impact of unexpected events on the prediction accuracy. Herein, we make use of a variable length Markov model to compare the usefulness of three alternatives to the hit and miss score: the Mean Absolute Error, the Ignorance Score, and the Brier score. We present an extensive evaluation of the methods on real data sets and a comprehensive comparison of the scoring methods.

Abstract
It is well known that the definition of bus schedules is critical for the service reliability of public transports. Several proposals have been suggested, using data from Automatic Vehicle Location (AVL) systems, in order to enhance the reliability of public transports. In this paper we study the optimum number of schedules and the days covered by each one of them, in order to increase reliability. We use the Dynamic Time Warping distance in order to calculate the similarities between two different dimensioned irregularly spaced data sequences before the use of data clustering techniques. The application of this methodology with the K-Means for a specific bus route demonstrated that a new schedule for the weekends in non-scholar periods could be considered due to its distinct profile from the remaining days. For future work, we propose to apply this methodology to larger data sets in time and in number, corresponding to different bus routes, in order to find a consensual cluster between all the routes. ©2010 IEEE.