This research paper addresses the challenge of maximizing crop yield through an intelligent approach in crop rotation using Q-learning and rule-based decision-making. The objective is to develop an adaptable crop rotation system that continuously monitors and refines the crop rotation sequence for enhanced efficacy and sustainability . Traditional crop rotation methods based on expert knowledge often fail to optimize crop yields, necessitating the use of computational models . The proposed approach integrates the Markov decision process (MDP) and Q-learning to determine an optimal crop rotation plan. A Q-learning approach is employed here to iteratively update Q-values and identify the best sequence of crops to be planted. The rewards are estimated based on yield values, providing a realistic assessment of the impact of crop rotation on crop yield. The effectiveness of the rotation sequence is evaluated over 50 years through year-wise yield assessments. Additionally, a rule-based decision-making process is introduced to modify the rotation sequence based on yield evaluations, ensuring continuous improvement in crop yield. The model is assessed using reward, Q-value, and yield, revealing improvements of 36.13%, 2.04%, and 35.73% in respective metrics. Simulation results confirm the proposed approach’s superiority over existing methods.

Cloud vendors (CVs) play an indispensable role in the development of IT sectors and industry 4.0. Many CVs evolve every day, and a systematic selection of these is becoming substantial for organizations. Literature studies have shown that multicriteria decision-making (MCDM) is a powerful tool for systematic selection. However, the major issue with the state-of-the-art models is that they do not effectively represent uncertainty. Moreover, the personalized selection of CVs based on user queries is not prominent in an MCDM context. In this paper, to circumvent these issues, a new decision framework is proposed that utilizes a generalized preference style called interval-valued probabilistic linguistic term set (IVPLTS). This preference style considers occurring probability values as interval numbers instead of a single precise value, which provides flexibility during preference elicitation. Initially, missing values are imputed systematically by using a case-based method. Then, the consistency of these preferences is checked using Cronbach's alpha coefficient, and the inconsistent preferences are repaired rationally by using an iterative method. A programming model is proposed for determining the weights of the evaluation criteria. Furthermore, Maclaurin symmetric mean (MSM) is extended to IVPLTS for aggregating preferences from each expert. The interval-valued probabilistic linguistic comprehensive (IVPLC) method is proposed for prioritizing CVs in a personalized manner. Finally, the framework's practicality is validated by using a case study of CV selection for an academic institution; strengths and weaknesses of the framework are conferred by comparison with extant CV selection models.

Diagrammatic reasoning (DR) problems are well known. However, solving DR problems represented in 4 × 1 Raven’s Progressive Matrix (RPM) form using computer vision and pattern recognition has not yet been tried. Emergence of deep learning techniques aided by advanced computing can be exploited to solve such DR problems. In this paper, we propose a new learning framework by combining LSTM and Convolutional LSTM to solve 4 × 1 DR problems. Initially, the elementary geometrical shapes in such problems are detected using a typical CNN-based detector. Next, relations of various shapes are analyzed and a high-level feature set is produced and processed in the LSTM framework. A new 4 × 1 DR dataset has been prepared and made available to the research community. We believe, it will be helpful in advancing this research further. We have compared our method with some of the existing frameworks that can be used for solving RPM-guided DR problems. We have recorded 18–20% increase in the average prediction accuracy as compared to the prior frameworks when applied to RPM-guided DR problems. We believe the CV research community will be interested to carry out similar research, particularly to investigate the feasibility of solving other types of known DR problems.

Abstract Surveillance camera usage has increased significantly for visual surveillance. Manual analysis of large video data recorded by cameras may not be feasible on a larger scale. In various applications, deep learning-guided supervised systems are used to track and identify unusual patterns. However, such systems depend on learning which may not be possible. Unsupervised methods relay on suitable features and demand cluster analysis by experts. In this paper, we propose an unsupervised trajectory clustering method referred to as t-Cluster. Our proposed method prepares indexes of object trajectories by fusing high-level interpretable features such as origin, destination, path, and deviation. Next, the clusters are fused using multi-criteria decision making and trajectories are ranked accordingly. The method is able to place abnormal patterns on the top of the list. We have evaluated our algorithm and compared it against competent baseline trajectory clustering methods applied to videos taken from publicly available benchmark datasets. We have obtained higher clustering accuracies on public datasets with significantly lesser computation overhead.

Synopsis of a long-duration video has many applications in intelligent transportation systems. It can help to monitor traffic with lesser manpower. However, generating meaningful synopsis of a long-duration video recording can be challenging. Often summarized outputs include redundant contents or activities that may not be helpful to the observer. Moving object trajectories are possible sources of information that can be used to generate the synopsis of long-duration videos. The synopsis generation faces challenges due to object tracking, grouping of the trajectories with respect to activity type, object category, and contextual information, and generating smooth synopsis according to a query. In this paper, we propose a method to generate meaningful and smooth synopsis of long-duration videos according to the users' query. We have tracked moving objects and adopted deep learning to classify the objects into known categories (e.g., car, bike, and pedestrians). We then identify regions in the surveillance scene with the help of unsupervised clustering. Each tube (spatiotemporal object trajectory) is represented by the source and the destination. In the final stage, we take a query from the user and generate the synopsis video by smoothly blending the appropriate tubes over the background frame through energy minimization. The proposed method has been evaluated on two publicly available datasets and our own surveillance datasets. We have compared the method with popular state-of-the-art techniques. The experiments reveal that the proposed method is superior to the existing techniques and it produces visually seamless video synopsis.

Abstract A quadratic minimum spanning tree problem determines a minimum spanning tree of a network whose edges are associated with linear and quadratic weights. Linear weights represent the edge costs whereas the quadratic weights are the interaction costs between a pair of edges of the graph. In this study, a bi‐objective rough‐fuzzy quadratic minimum spanning tree problem has been proposed for a connected graph, where the linear and the quadratic weights are represented as rough‐fuzzy variables. The proposed model is formulated by using rough‐fuzzy chance‐constrained programming technique. Subsequently, three related theorems are also proposed for the crisp transformation of the proposed model. The crisp equivalent models are solved with a classical multi‐objective solution technique, the epsilon‐constraint method and two multi‐objective evolutionary algorithms: (a) nondominated sorting genetic algorithm II (NSGA‐II) and (b) multi‐objective cross‐generational elitist selection, heterogeneous recombination, and cataclysmic mutation (MOCHC) algorithm. A numerical example is provided to illustrate the proposed model when solved with different methodologies. A sensitivity analysis of the example is also performed at different confidence levels. The performance of NSGA‐II and MOCHC are analysed on five randomly generated instances of the proposed model. Finally, a numerical illustration of an application of the proposed model is also presented in this study.

Transportation management is one of the key success factors to keep an organization competitive, sustain its growth pace, and raise profits not only at a local but also a global scale. Therefore, planning and designing a transport system are prerequisite and vital topics for achieving these goals. In this paper, a multiobjective multi-product solid transportation problem (MOMPSTP) under uncertainty is formulated and solved by two different methods of multiobjective optimization problems (MOPs). The system parameters namely unit transportation cost, availability of products at source points, demands of products at destinations and the capacity of transportation mode all are taken as rough fuzzy variables (RFVs). A chance constraint programming model for MOP with RFVs is developed in order to obtain satisfactory solutions when decision makers (DMs) aim to optimize multiple objectives (cost, time, profit, etc.) simultaneously. For given credibility (Cr) and trust (Tr) levels of RFVs, Cr-Tr constraint programming technique is used to reduce the uncertain transportation problem into equivalent deterministic form. Two classical solution techniques-weighted sum method (WSM) and ideal point method (IPM) are utilized to solve the problem. Finally, a numerical example is provided to illustrate the usefulness of our proposed model and then a sensitivity analysis is performed to verify different solutions due to different level of satisfaction.

Due to the advancement of camera hardware and machine learning techniques, video object tracking for surveillance has received noticeable attention from the computer vision research community. Object tracking and trajectory modeling have important applications in surveillance video analysis. For example, trajectory clustering, summarization or synopsis generation, and detection of anomalous or abnormal events in videos are mainly being exploited by the research community. However, barring one research work (which is almost a decade old), there is no recent review that emphasizes the use of video object trajectories, particularly in the perspective of visual surveillance. This paper presents a survey of trajectory-based surveillance applications with a focus on clustering, anomaly detection, summarization, and synopsis generation. The methods reviewed in this paper broadly summarize the abovementioned applications. The main purpose of this survey is to summarize the state-of-the-art video object trajectory analysis techniques used in the indoor and outdoor surveillance.

The papers in this special section provide a series of high quality papers which significantly contribute to new theories, applications, and algorithms about uncertain multicriteria decision making. It not only reports recent significant developments but also highlights potential, growing research directions, and future trends, which will benefit researchers in studying theories, applications, and algorithms about uncertain multicriteria/ multiobjective/multiattribute decision making.

Abstract This paper investigates the uncertain maximum flow of a network whose capacities are random fuzzy variables. We have developed the expected value model (EVM) and the chance‐constrained model (CCM) for maximum flow problem (MFP) under random fuzzy environment and formulated their crisp equivalent models. To solve these models, we have proposed a varying population genetic algorithm with indeterminate crossover (VPGAwIC). In VPGAwIC, selection of a chromosome depends on its lifetime. An improved lifetime allocation strategy (iLAS) has also been proposed to determine the lifetime of the chromosome. The ages of the chromosomes are defined linguistically as Young, Middle, and Old, which follow some uncertainty distributions. The crossover probability is indeterminate, and it depends on the ages of the parents, which is defined by an uncertain rule base. The number of offspring, generated from a population of parents, is determined by the reproduction ratio. The population is updated in 2 ways: (i) All the chromosomes with ages greater than their lifetimes are discarded from the population, and (ii) the offspring are combined with their parents for the next generation. The proposed VPGAwIC is compared with the genetic algorithm developed by Gen, Cheng, and Lin (2008) for maximum flow problem. Wilcoxon signed‐rank test has been performed to show the superiority of the proposed VPGAwIC.

Abstract This paper presents a novel Multiple Criteria Decision Making methodology for assessing and prioritizing medical tourism destinations under uncertainty. A systematic evaluation and assessment approach is proposed by incorporating analytic hierarchy process and multi‐attributive border approximation area comparison methods in the rough environment. Rough number is used to aggregate individual judgements of decision makers and express their true perception to handle vagueness without any prior information. Rough analytic hierarchy process analyses the relative importance of criteria based on their preferences given by experts, whereas rough multi‐attributive border approximation area comparison evaluates the alternative sites based on the criteria weights. A case study of prioritizing different sites (cities) in India for medical tourism services is shown to demonstrate the applicability of the proposed method. Among different criteria “quality of infrastructure of healthcare institutions” is observed to be the most important criteria in our analysis, followed by “supply of skilled human resources and new job creations” and “Chennai” is found to be the best medical tourism site in India. Finally, a comparative analysis and validity testing of the proposed method are elaborated, and the methodology provides a standard for select medical tourism sites on the basis of different criteria.

Uncertain calculus is a branch of mathematics that deals with the differentiation and integration of functions of uncertain processes. This paper investigates the Yao process defined by the Yao integral and extends the uncertain calculus on the Yao process. Some important results are developed for finite variation and linearity of integral. Moreover, this paper proposes a concept of the multifactor Yao process and also gives a fundamental theorem for it.

A noticeable progress has been found in decision making problems since the introduction of soft set theory by Molodtsov in 1999. It is found that classical soft sets are not suitable to deal with imprecise parameters whereas fuzzy soft sets (FSS) are proved to be useful. Use of intuitionistic fuzzy soft sets (IFSS) is more effective in environment, where arguments are presented using membership and non-membership values. In this paper we propose an algorithmic approach for multiple attribute group decision making problems using interval-valued intuitionistic fuzzy soft matrix (IVIFSM). IVIFSM is the matrix representation of interval-valued intuitionistic fuzzy soft set (IVIFSS), where IVIFSS is a natural combination of interval-valued intuitionistic fuzzy set and soft set theory. Firstly, we propose the concept of IVIFSM. Then an algorithm is developed to find out the desired alternative(s) based on product interval-valued intuitionistic fuzzy soft matrix, combined choice matrix, and score values of the set of alternatives. Finally, a practical example has been demonstrated to show the effectiveness of the proposed algorithm.

In this paper, we have proposed a fuzzy interval time series model using a new strategy to replace the conventional defuzzification step, where genetic algorithm has been used to optimize the interval parameters and neural network has been used to learn the trend of the time series. First order fuzzy time series with equal time interval has been used on two data sets, enrollments of the University of Alabama and gold exchange traded fund. We compare the proposed model with two other existing models. The results of the comparisons show that the proposed model performs better.

Efficiency assessment and determining optimal low risk financial institution for monetary aid is among the crucial issues which multinational companies are facing in risk oriented supply chain. Supported by fuzzy-soft tools we present a hybrid model for assessing the uncertain fragile and risky structure of the financial sector which has direct impact on processing stages of supply chain network. By integrating Trapezoidal Interval based Type-2 fuzzy soft sets with Extended Type-2 TOPSIS we propose a first-hand approach to multi-criteria decision making problem. Using linguistic rating system, weights of risk criteria are assessed based on trapezoidal interval type-2 fuzzy soft sets. Through Type-2 TOPSIS, the paper proposes Euclidean distance between ideal solutions and relative degree of closeness as evaluative standard for ranking the financial alternatives. Finally, the largest six commercial financial institution of Indian Banking Sector are examined and validated on basis of five risk financial criteria's. The proposed method furnishes the solution of decision problem with less computational effort.

Hypertension is called the silent killer because it has no symptoms and can cause serious trouble if left untreated for a long time. It has a major role for stroke, heart attacks, heart failure, aneurysms of the arteries, peripheral arterial diseases, chronic kidney disease etc. An intelligent and accurate diagnostic system is mandatory for better diagnosis and treatment of hypertension patients. This study develops a fuzzy expert system to diagnose the hypertension risk for different patients based on a set of symptoms and rules. Next we design a neuro fuzzy system for the same set of symptoms and rules using three different types of learning algorithms which are Levenberg-Marquardt (LM), Gradient Descent (GD) and Bayesian Resolution (BR) based learning functions. Then this paper presents a comparative study between fuzzy expert system (FES) and feed forward back propagation based neuro fuzzy system (NFS) for hypertension diagnosis. This paper also presents a comparison among the learning functions (LM, GD and BR) where Levenberg-Marquardt based learning function shows its efficiency over the others. Comparison between FES and NFS shows the effectiveness of using NFS over FES. Here, the input data set has been collected from 10 patients whose ages are between 20 and 40 years, both for male and female. The input parameters taken are age, body mass index (BMI), blood pressure (BP), and heart rate. The diagnosis process, linguistic variables and their values were modeled based on expert's knowledge and from existing database.