International Journal of Information and Decision Sciences (19 papers in press)
by Duangruthai Voramontri, Leslie Klieb
Abstract: The goal of this paper is to research empirically the role of social media in consumers decision-making process for complex purchases - those characterized by significant brand differences, high consumer involvement and risk, and which are expensive and infrequent. The model uses the information search, alternative evaluation, and purchase decision stages from the classical EBM model. A quantitative survey investigates up to what degree experiences are altered by the use of social media. Results show that social media usage influences consumer satisfaction in the stages of information search and alternative evaluation, with satisfaction getting amplified as the consumer moves along the process towards the final purchase decision and post-purchase evaluation. The research was done among internet-savvy consumers in South-East Asia, and only considered purchases that were actually made by consumers, not including searches that were abandoned.
Keywords: social media; consumer decision-making; EBM model; EKB; information search; complex purchase.
The impact of using new significant reference point with TOPSIS methods: study and application
by Zhor CHERGUI
Abstract: In this paper, the impact of Pareto optimality concept on revised TOPSIS method is studied. In particular, we study theoretically the cases in which a preference relation changes when delimiting the choice of the best alternative(s) in an efficient restrictive area. In order to define the most reliable approach a comparative study is established. On this basis, an accurate new method called TOPSIS Nadir is introduced. Furthermore, an adaptation for Interval data area is carried out in which we discuss some forms of normalization. By following the same steps of the TOPSIS methods for Group Decision Makers, we develop and compare two new procedures.
Keywords: Group Decision Makers; TOPSIS methods; Reference points; Crisp data & Interval data; Forms of normalization.
Spatiotemporal Assessment of Water Quality in the Distribution Network of City of Sharjah, UAE
by Maruf Mortula, Kazi Fattah, Tarig Ali, Alaeldin Idris, Mayyada AlBardan
Abstract: Maintaining a healthy water distribution network (WDN) is key to providing good quality services to the consumers in a sustainable manner. WDN in the City of Sharjah, United Arab Emirates (UAE), has more than 3000km of pipeline and receives water from different sources. Understanding the variation of water quality over time is critical to appropriate management. The objective of this paper was to assess the variability of the water quality in Sharjah WDN. Monitoring data for residual chlorine, iron, and fluoride were collected from 46 different locations throughout the distribution system. Graphical and GIS-based analyses were conducted to understand the temporal (for three source waters and three locations in the WDN) and spatial variability (all locations) of water quality throughout the distribution network. Temporal variations indicated seasonal water quality variations throughout the three-year period (2012-2014). The spatial variability indicated that the old part of the city was susceptible to water quality degradation.
Keywords: water distribution network; water quality; infrastructure integrity; geographic information system; spatial variability; infrastructure management.
A non-stationary NDVI time series modelling using Triplet Markov Chain
by Ali Ben Abbes, Mohamed Farah, Imed Riadh Farah, Vincent Barra
Abstract: Nowadays, vegetation monitoring using remotely sensed data is an
important far-reaching real-world issue.The main purpose of this study is to
build a Triplet Markov Chain (TMC) to model and analyse vegetation dynamics
on large scales using non-stationary Normalised Difference Vegetation Index
(NDVI) time series.TMC is a generalisation of Hidden Markov Models (HMMs),
which have been widely used to represent Satellite Time Series Images but
which they proved to be inefficient for non-stationary data. The TMC model
proposed in this paper overcomes this limit by adding an auxiliary process
which allows modelling non-stationarity. In order to assess the performance of
the proposed model, experimentation is carried out using Moderate Resolution
Imaging Spectroradiometer (MODIS) NDVI time series of the northwestern
region of Tunisia. The TMC model is compared to standard HMM and Seasonal
Auto Regressive Integrated Moving Average model (SARIMA) and proved to
achieve the best performance with an overall accuracy prediction rate of 92.8%
and a kappa coefficient of 0.885.
Keywords: NDVI Time Series; Vegetation Dynamics; TMC; HMM; Non-Stationarity; Remote Sensing.
A novel architecture based on fuzzy cognitive maps and holonic systems for decision making in a cooperative context
by Asma Maziz, Nacereddine Zarour
Abstract: Ensuring consistency and good decision making is one of the most topical problems in an information system; it becomes more difficult in a cooperative context. In this paper, we propose an architecture based on fuzzy cognitive maps (FCM) tool and holonic multi-agent paradigm that enhance the decision making process in cooperative information system (CIS). Furthermore, the concept of ontology is used for semantically enrich our architecture. We modeled each sub-CIS by a holonic agents where everyone used a FCM for a more precise analysis of complex dynamic system decisions. This group will try to make a collective decision to solve any given distributed problem. To put our approach into practice, we considered road safety field to see how to educate people in order to reduce the fatal accidents number. Finally, we validated our proposition through experiments to show how it improves the decision making process in a cooperative context.
Keywords: Decision Support; Cooperative Information Systems; Fuzzy Cognitive Maps; Multi-agent Systems; Holonic systems; Ontology; Road safety.
Decision Support for Nutrition Management of Grapes using Ontology based on Decision Trees
by Archana Chougule, Vijay Kumar Jha, Debajyoti Mukhopadhyay
Abstract: For any decision support system, having meaningful, up-to-date, interoperable and consistent knowledge base is important. Ontologies can be sued for knowledge semantics and knowledge sharing. Hence ontologies are getting more importance these days as heterogeneous integrated systems are used in almost all areas. Ontology gets evolved with increase in domain knowledge of experts. Change management of ontology is must to keep consistency of knowledge base. This paper demonstrates use of decision tree for ontology building and evolution. Detail algorithm for extending ontology from decision tree is discussed in the paper. For decision support using knowledge in ontology, ontology reasoning is used. Semantic web rule language is the technique used for ontology reasoning. Accuracy of decision support depends on strength and correctness of inference logic. Paper describes how accuracy of decision support improves with semi-automated construction of SWRL rules. The approach is validated with example of nutrition management system for grapes.
Keywords: Decision support; ontology evolution; decision tree; semantic web rule language; grapes; nutrition management.
A decision making methodology for material selection using Genetic Algorithm
by Elyas Abbasi Jannatabadi, Masoud Goharimanesh, Ali Jahan, Ali Akbar Akbari
Abstract: Material selection is a challenging task for designers due to the immense number of different materials available today. Choosing the right materials plays an important role in numerous engineering applications because an inappropriate selection of materials can significantly affect the performance of the final product. As a result, a number of techniques have been proposed to select materials in the engineering design process. However, most of the proposed systems are knowledge intensive and cannot deal with the situation where the information of weight criteria is incomplete or unknown. So, in this paper a logical approach is presented for choosing an optimal material by employing the Genetic algorithm. The proposed material selection procedure reduces the personal bias for assigning the weight of different attributes. Seven examples are included to demonstrate the applicability of the suggested approach. The findings of this work provide the insights for further researches on more complicated design problems such as simultaneous material selection and geometry optimization.
Keywords: Materials selection; Genetic Algorithm; multiple criteria analysis; multi criteria decision making; weighting factors; Ranking.
Efficient Evacuation in a Multi-Exit Environment: An Agent-based Decision Support Model
by Kashif Zia, Dinesh Saini, Arshad Muhammad
Abstract: A majority of research work carried out in crowd evacuation rely on simulation due to non-availability of real and realistic trial data. In this paper, an agent-based simulation study of an evacuating crowd is presented. The model is based on the microscopic behavioral rules formulated through small-scale empirical evidence in conjunction with crowd behavioral theories. In particular, the study focus on the possibility of efficient evacuation from the environment with limited perceptions. Extending Moore's neighborhood model, local congestion avoidance mechanism capable of detecting the relative displacement and orientation of the all the individuals in its neighborhood is considered. Other strategies based on exit capacity and exit population are also modeled and tested. A probabilistic exit selection strategy is also designed that considers a sensitivity of an exit as a deciding factor. The simulation results show that the enhanced exit selection strategies make the proposed system more robust and increase the evacuation efficiency substantially.
Keywords: Crowd evacuation; Decision support model; Multi-exit efficiency; Agent-based modeling; NetLogo simulation.
An Integrated Fuzzy Delphi and Fuzzy Inference System for Ranking the Solutions to Overcome the Supply Chain Knowledge Flow Barriers
by Vishal Bhosale, Ravi Kant
Abstract: Supply Chain (SC) is assumed as a leading operations strategy in both manufacturing and service organizations. With rapid change and competition in the SC, knowledge is recognized as an important source of competitive advantage. In todays, business word, organizations should manage the SC knowledge to stay ahead in the competition. However, evidence suggests that there are several SC knowledge flow barriers (SCKFBs) which obstruct knowledge flow in SC. The purpose of this research is to identify SCKFBs and propose solutions to overcome the SCKFBs. In this study an example of Indian automobile brake manufacturing organization is presented to exemplify the use of the proposed framework for SCKFBs and solutions to overcome them. The weight of major SCKFBs is calculated by fuzzy Delphi (FD) method and ranking of the solutions of SCKFBs is evaluated using the fuzzy inference system (FIS) method. To overcome SCKFBs, the top rank solutions obtained are visible technologies that offer real-time customer and demand knowledge, building strategic relationships with SC partners and induce mutual trust with SC partners. Visible technologies offer real-time customer demand data; which enable SC partners to maximize operational efficiencies and enhance customer value creation. The extensive list of solutions of SCKFBs facilitates an organization to focus on higher ranked solutions and improve policies to implement them. This study may be one of the first to bring together a large range of SCKFBs on the same platform, and an attempt to give solutions for its barriers in order to put SC to the next level using integrated FD-FIS methods.
Keywords: Supply chain; knowledge flow; Fuzzy Delphi; Fuzzy inference system.
Studying the Effect of Community Structure for Seed Selection in an Influence Model
by Carolina Xavier, Vinícius Vieira, Alexandre Evsukoff
Abstract: This paper presents a study of influence spreading in real complex networks which shows that community structure in networks can be used to guide the selection of seed nodes to spread information and ideas over the network. The results obtained by the application of the methodology to a set of benchmark networks suggest that the distribution of seeds between the central nodes of the networks communities can increase the range of information spreading when compared to alternative methods using central nodes as seeds considering only the global context of the network.rn
Keywords: Influence maximization; seed selection; communities; Spreading Activation Model.
Enhancing the performance of sentiment analysis task on product reviews by handling both local and global context
by Bagus Setya Rintyarna, Riyanarto Sarno, Chastine Fatichah
Abstract: Commonly, product review analysis includes extracting sentiment from product documents. The contextual aspect contained in a review document has potential to improve results obtained by the sentiment analysis task of product reviews. In this regard, this paper proposes an approach that takes into account both local and global context. The main contribution of this work is threefold. Firstly, local context is defined and the graph-based Word Sense Disambiguation (WSD) method is extended to deal with this contextual issue. The method is aimed at assigning the correct sense of a word in the context of a sentence, which means choosing the correct sentiment value of a word with respect to the context. Secondly, global context is defined for addressing contextual issues related to the specific domain of a review document, which can affect the sentiment value of the words contained in it. To address the global context issue, an improved SentiCircle-based method is used and a similarity-based technique is provided to select the pivot word. This method can be employed to assign sentiment value at sentence level. Thirdly, a weighted mean-based strategy to determine sentiment value at document level is presented. Several experiments were conducted to assess the proposed method and compare it with a baseline method. Overall, the proposed method outperformed the baseline method in almost all performance evaluation measures (precision, recall, F-measure and accuracy).
Keywords: sentiment analysis; local context; global context; word sense disambiguation; SentiCircle.
A new approach agent-based for distributing association rules by business to improve decision process in ERP systems
by Merouane Zoubeidi, Okba Kazar, Saber Benharzallah, Nadjib Mesbahi, Abdelhak Merizig, Djamil Rezki
Abstract: In the last decade, the distributed computing plays an important role in the Data Mining process, it helps to make systems scalable and it is important to develop mechanisms that distribute the workload among several sites in a flexible way also the acronym ERP stands for enterprise resource planning. It refers to the systems and software packages used by organizations to manage day-by-day business activities, ERP systems are designed for the defined data structure (schema) that usually has a common database. In addition, Data Mining is a technology that purposes to promote information and knowledge extraction from a large database. In this paper, we present a collaborative multi-agent based system for association rules mining from distributed databases. In our proposed approach
Keywords: Enterprise Resource Planning (ERP); Multi-Agents system (MAS); Data Mining associate rules; JADE; WEKA.
A rule-based approach for dynamic Analytic Hierarchy Process decision making
by Yun-ning Liu, Shiow-yang Wu
Abstract: The Analytic Hierarchy Process (AHP) is widely used in many multi-criteria decision-making problems and has been successfully applied to many practical cases. However, the AHP process is time-consuming and the decision model is not agile enough for fast changing environment. To overcome this weakness, we develop a rule-based approach for dynamic AHP decision-making in changing environment. We analyze critical factors in the AHP decision process under uncertainty and propose to encode expert knowledge for change handling using Event-Condition-Action rules. We propose a theorem and associated method to determine the change in ordering of decision alternatives based on Event-Condition-Action rule-induced weight updates. We demonstrate the effectiveness of our approach using a case study of the supplier selection decision making task of the steel and iron industry in Taiwan. The study shows that our mechanism can effectively reach the same level of decision quality as expert decision maker(s).
Keywords: Dynamic rule-based AHP; Two Criteria Update Impact Analysis; Steel and Iron Industry; Comparison Matrix.
Quality of Service based Service Selection in Smart Parking
by Shiksha Singh, Rohit Kumar Tiwari
Abstract: Smartness in the existing environment is required for overall growth of any country. Government is putting tremendous effort and a huge sum of money for making cities smart to achieve smartness. A very first asset that needs to be made smart is smart parking system to avoid the traffic congestion. There are many service providers which are offering smart parking services, but there is no Quality of Service (QoS) framework available so far. So, for a customer point of view it is very hard to select the best service provider and gain maximum satisfaction. So, in this paper, we have designed a QoS framework which consists of thirty-three metrics to evaluate a smart parking service. These parameters are identified from the user as well as vendors perspective and helps to select better service provider. We have also proposed multi criteria decision making (MCDM) approach TOPSIS to select best smart parking service provider based on identified QoS. We have demonstrated our approach with the help of case studies.
Keywords: Quality of Services; Smart City; Smart Parking System; Internet of Things; MCDM; TOPSIS.
A new framework using biform game for cost optimization of distribution networks
by Salma MOUATASSIM, Ahmed Haroun SABRY, Mustapha AHLAQQACH, Jamal BENHRA
Abstract: The present work focuses on the demand decision making problem for regional distribution centers sharing the same product families. Each center orders quantities to be distributed from production units. Our approach suggests a biform game to maximize the benefits of each center and minimize the end of cycle market induced supply to demand deviations. We start by an independent demand forecasting under uncertainty. Once the demand is met, the centers enter a collaboration phase where coalitions are created and products are exchanged, in order to achieve the core stability of the actual game. If not met, we try to achieve the same objectives using individual rationality through an adapted approach based on Shapley value analysis for each possible coalition.
Keywords: game theory; forecasting; Shapley value; collaboration; biform game; cost allocation; coalitions; core stability.
Determining Optimal Replanting Rate in Palm Oil Industry, Malaysia: A System Dynamics Approach Optimal Policy Search in Oil Palm Plantation Feedback Loops using System Dynamics Optimization
by Mohd Zabid M Faeid, Norhaslinda Zainal Abidin, Shri Dewi Applanaidu
Abstract: One of the important factor that contribute to the stagnant growth of Malaysias crude palm oil production is the accumulation of ageing oil palm plantation area. Given the scarcity of new plantation area in Malaysia, it is very important that an optimal replanting rate has to be determined to decrease the accumulation of ageing area. The mature area of oil palm plantation has to be increased in order to obtain higher crude palm oil production. The main aim of this study is to determine an optimal replanting rate for oil palm industry in Malaysia.. This study compared the trend results of fresh fruit bunches yield using baserun system dynamics and system dynamics with optimization analysis.The findings indicate that the proposed optimal replanting rate based on system dynamics analysis revealed the maximum production in term of the fresh fruit bunch yield by year 2050 compared to baserun scenario analysis. Specifically, findings from optimization analysis recommended the replanting rate at 278,189 hectare annually. This practise will ensure the continuous supply of palm oil withoit significant distruption if high replanting rate has been implemented.. Findings from this study will be useful to the policy makers in palm oil industry and palm oil planters in assisting them to plan the appropriate planting strategy to maximize fresh fruit bunches yield.
Keywords: Fresh fruit bunches yield; Oil palm plantation; Optimal replanting rate; System dynamics optimization.
A SWITCHING HYBRID MOBILE RECOMMENDER SYSTEM FOR TOURISTS
by Bolanle Ojokoh, Idorenyin Amaunam
Abstract: Recommender systems are meant to give recommendations by receiving and analyzing feedbacks from users. These feedbacks could be obtained through ratings or likes. Nevertheless, a problem arises when there are no feedbacks. This can happen because: the system offering the recommendations is new (and has obtained no ratings), the user is new (and has not rated the system), or the items for recommendations are new (and have not been rated). This problem is termed the cold-start problem. Also, most recommender systems do not offer preferred recommendations to users. They rely mostly on the ratings of other users to give recommendations to the active user. These in most cases may not be the users choice. This paper thus, proposes a switching feature-based model that leverages the need of both new and existing users for recommendation of tourist locations. Recommendations to new users are implemented with Bayesian algorithm on supplied demographic data. For existing users, the system switches to the collaborative filtering subsystem, where the user inputs the location facility he/she wants. The user also has the privilege of changing location type. Recommendation results are produced through the appropriate algorithm and offered based on the items in the database. The model was validated with discounted cumulative gain, precision, and recall. A comparative analysis with some existing systems showed lower mean absolute error. Also, experimental results obtained through questionnaire distributed to general users, users from the computer science domain as well as experts in the tourism, showed the effectiveness of the proposed techniques.
Keywords: Bayesian algorithm; Conditional Probability Table (CPT); Cold-start; Mobile app; Recommender system.
Special Issue on: DASA'16 Decision and Logistics
Modeling Time Complexity of Micro-Genetic Algorithms for Online Traffic Control Decisions
by Ghassan Abu-Lebdeh, Kenan Hazirbaba, Omer Mughieda, Bassant Abdelrahman
Abstract: Genetic Algorithms are especially effective optimization tool when (near) optimal solutions for traffic control decisions for large-scale combinatorial problems are sought in real time. Optimizing online traffic control in urban networks is such a problem where online control decisions are recurrently sought based on real-time traffic information input. Limited time is available to reach near optimal solutions hence the best-thus-far solution (BTFS) is often the best one can do. It is thus critical that the best of BTFSs is identified. For that, offline evaluation is necessary to ensure appropriate combination of GA parameters and operators are selected for the real time online implementation in a real world setting. This paper describes an experimental approach to test the suitability of micro-Genetic Algorithms (m-GAs) to solve very large combinatorial traffic control problems and establishes relationships between time to convergence and problem size. A discrete time dynamical traffic control problem with different sizes and levels of complexity was used as a test-bed. The results showed that m-GAs can tackle computationally demanding problems. Upon appropriately sizing the m-GA population, the m-GA converged to a near-optimal solution in a number of generations equal to the string length. The results also demonstrated that with the selection of appropriate number of generations, it is possible to get most of the worth of the theoretically optimal solution but with only a fraction of the computation cost. The results showed that as the size of the optimization problem grew exponentially, the time requirements of m-GA grew only linearly thus making m-GAs especially suited for optimizing large scale and combinatorial problems for on-line optimization.
Keywords: On-line Traffic Control Decisions; Genetic Algorithms; Sustainable Transport; Genetic Algorithms Convergence.
Special Issue on: DASA'16 Decision and Logistics
Decision Aid System for Omani medical herb leaves recognition using computer vision and artificial intelligence
by Majed Bouchahma, Mohsin Al- Balushi, Sheikha Al-Housni, Hamood Al-Wardi
Abstract: Herbs have been widely used in food preparation, medicine and cosmetic industry. Knowing which herbs to be used would be very critical in these applications. This research aims to define a method to classify the herbs plants based on their leaves colors and shapes. An Open Source Decision Aid System is designed and developed especially for helping scientist. The proposed system employs artificial and image processing techniques to perform recognition on a number of Omani species of medical herbs.
Keywords: Decision Aid System; medical herbs; computer vision; artificial intelligence;SURF.