A large number of software reliability growth models (SRGMs) have been proposed during the past 30 years to estimate software reliability measures such as the number of remaining faults, software failure rate, and software reliability. Selection of an optimal SRGM for use in a particular case has been an area of interest for researchers in the field of software reliability. Tools and techniques for software reliability model selection found in the literature cannot be used with high confidence as they use a limited number of model selection criteria. For the first time, we developed a deterministic quantitative model based on a distance based approach (DBA) method, then applied it for evaluation, optimal selection, and ranking of SRGMs. DBA recognizes the need for relative importance of criteria for a given application, without which inter-criterion comparison could not be accomplished. It requires a set of model selection criteria, along with a set of SRGMs, and their level of criteria for optimal selection; and it successfully presents the results in terms of a merit value which is used to rank the SRGMs. We use two distinct, real data sets for demonstration of the DBA method. The result of this study will be a ranking of SRGMs based on the Euclidean composite distance of each alternative to the designated optimal SRGM.

With the widespread adoption of renewable energy sources in the smart grid era, there is an utmost requirement to develop prediction models that can accurately forecast solar irradiance. The stochastic nature of solar irradiance considerably affects photo-voltaic (PV) power generation. Since weather conditions have a high impact on solar irradiance; therefore, we need weather-conscious forecasting models to boost predictive accuracy. Although Recurrent Neural Networks (RNNs) has shown considerable performance in time-series forecasting problems, its sequential nature prohibits parallelized computing. Recently, architectures based on self attention mechanism have shown remarkable success in natural language programming (NLP), while being computationally superior. In this paper, we propose an RSAM (Robust Self-Attention Multi-horizon) forecasting architecture, which mainly works in two parts: First, multi-horizon forecasting of solar irradiance using multiple weather parameters; Second, prediction interval analysis for model robustness using quantile regression. A self-attention based Transformer model belonging to the family of deep learning models has been utilized for multi-variate solar time-series forecasting. Using the National Renewable Energy Laboratory (NREL) benchmark datasets of two different sites, we demonstrate that the proposed approach exhibit enhanced performance in comparison to RNN models in terms of RMSE, MAE, MBE and Forecast skill at each forecasted interval.

With the advancement of wireless communication, internet of things, and big data, high performance data analytics tools and algorithms are required. Data clustering, a promising analytic technique is widely used to solve the IoT and big data based problems, since it does not require labeled datasets. Recently, meta-heuristic algorithm have been efficiently used to solve various clustering problems. However, to handle big data sets produced from IoT devices, these algorithm fail to respond within desired time due to high computation cost. This paper presents a new meta-heuristic based clustering method to solve the big data problems by leveraging the strength of MapReduce. The proposed methods leverages the searching potential of military dog squad to find the optimal centroids and MapReduce architecture to handle the big data sets. The optimization efficacy the proposed method is validated against 17 benchmark functions and the results are compared with 5 other recent algorithms namely, bat, particle swarm optimization, artificial bee colony, multiverse optimization, and whale optimization algorithm. Further, the parallel version of the proposed method is introduced using MapReduce (MR-MDBO) for clustering the big datasets produced from industrial IoT. Moreover, the performance of MR-MDBO is studied on 2 benchmark UCI datasets and 3 real IoT based datasets produced from industry. The F-measure, and computation time of the MR-MDBO is compared with the 5 other state-of-the-art methods. The experimental results witness that the proposed MR-MDBO based clustering outperforms the other considered algorithms in terms of clustering accuracy and computation times.