Uncertainty Analysis in Group Decisions through Interval Ordinal Priority Approach

2023 | Group Decision and Negotiation | Citations: 0

Authors: Mahmoudi, Amin; Javed, Saad Ahmed

Abstract: In multiple criteria decision-making (MCDM) problems, ranking alternatives is us ... Expand

Abstract: In multiple criteria decision-making (MCDM) problems, ranking alternatives is usually the end. However, in real life, it is rarely an end in itself and serves as a means to an end. Further, in real-life problems, the selection of alternatives is usually made with the aid of unique experts, and thus minimizing the influence of the uncertainty of their subjective judgements on the optimality of the decisions is a critical issue. The current study proposes a novel Interval Ordinal Priority Approach to objectively solve these and other issues by allowing uncertainty analysis and quantification. The study argues that when the model’s input contains uncertainty (even if represented by crisp numbers), expecting the output to be free from uncertainty is an unrealistic conjecture. Therefore, unlike the conventional MCDM models producing crisp weights, the proposed approach yields interval weights with the length of the interval representing the uncertainty (inconsistency among the experts’ judgements). Also, instead of resorting to the subjective measurement of thresholds to qualify or disqualify a set of inputs based on the degree of uncertainty, a novel objective measure of threshold is put forward. The validity of the proposed method is demonstrated through illustrative examples and comparative analysis. Later, the study is concluded with the implications for real-world decision-making. Collapse

Topics: decision making decision support system

Methods: descriptive statistic cross sectional research statistical hypothesis test trade-off analysis method Grey relational analysis

Theories: mathematical theory

Model identification for ARMA time series through convolutional neural networks

2021 | Decision Support Systems | Citations: 0

Authors: Tang, Wai Hoh; Röllin, Adrian

Abstract: We use convolutional neural networks for model identification in ARMA time serie ... Expand

Abstract: We use convolutional neural networks for model identification in ARMA time series models, where our networks are trained on synthetic data with known ground truths. Comparing the performance of these networks with traditional likelihood-based methods, in particular the Akaike and Bayesian Information Criteria, we are able to show that when it comes to statistical inference on ARMA orders, neural networks can significantly outperform likelihood-based methods in terms of accuracy and, by orders of magnitude, in terms of speed. We also observe improvements in terms of time series forecasting. Our approach shows the feasibility of using artificial neural networks for statistical inference in situations where classical likelihood-based methods are difficult or costly to implement. Collapse

Methods: time series analysis longitudinal research autoregressive moving average model artificial neural network residual neural network

Theories: mathematical theory

Triggers of Social Network Collapse

2020 | Information Systems Management | Citations: 0

Authors: Belik, Ivan

Abstract: Modern social networks are characterized by highly connected structures that ref ... Expand

Abstract: Modern social networks are characterized by highly connected structures that reflect a variety of social interrelations. In the given research, we focus on the quantitative analysis of social networks in terms of their sustainability. Based on the symbiosis of game theory, network analysis, and decision theory, we introduce an interdisciplinary approach for the detection of agents that have the potential social power to trigger the dissolution processes in networks. Collapse

Topics: social network Facebook MySpace shapley additive explanation Google+

Methods: computational algorithm game theoretical model experiment network analysis social network analysis

Theories: decision theory game theory mathematical theory

BUSINESS VALUE OF SMART CONTRACT: CASE OF INVENTORY INFORMATION DISCREPANCIES

2020 | Americas Conference on Information Systems | Citations: 0

Authors: Padalkar, Nakul; Sheikh-Zadeh, Alireza; Song, Jaeki

Abstract: Firms are increasingly interested in Blockchain Smart Contracts as a solution f ... Expand

Abstract: Firms are increasingly interested in Blockchain Smart Contracts as a solution for the visibility of the digital supply chain. Blockchain can help realize the cost reductions by providing a "single version of the truth" for a firm and its trading partners. By sharing important information such as inventory levels, manufacturing performance and operations indicators, and order and shipment information, firms can eliminate the delays and uncertainties in the information that contributes to "the bullwhip effect" and inflates required buffer stock. This paper focuses on the impact of blockchain in an inventory operation to cope up with information discrepancies. We present and compare the cost differences between an existing technology (like EDI) and blockchain. Using technologies like blockchain and smart contracts will enable a more transparent, sustainable, and resilient supply chain. Collapse

Topics: blockchain smart contract electronic data interchange information exchange supply chain management

Methods: mathematical model

Theories: mathematical theory economic production theory

Reverse Engineering Integrated Circuits Using Finite State Machine Analysis

2017 | HICSS | Citations: 0

Authors: Smith, Jessica; Oler, Kiri; Miller, Carl; Manz, David

Abstract: Due to the lack of a secure supply chain, it is not possible to fully trust the ... Expand

Abstract: Due to the lack of a secure supply chain, it is not possible to fully trust the integrity of electronic devices. Current methods of verifying integrated circuits are either destructive or non-specific. Here we expand upon prior work, in which we proposed a novel method of reverse engineering the finite state machines that integrated circuits are built upon in a non-destructive and highly specific manner. In this paper, we present a methodology for reverse engineering integrated circuits, including a mathematical verification of a scalable algorithm used to generate minimal finite state machine representations of integrated circuits. Collapse

Topics: computer hardware systems implementation

Methods: computational algorithm experiment

Theories: mathematical theory

A Decision Rule Aggregation Approach to Multiple Criteria-Multiple Participant Sorting

2012 | Group Decision and Negotiation | Citations: 0

Authors: Chen, Y.; Kilgour, D. Marc; Hipel, Keith W.

Abstract: A system that aggregates case-based linguistic decision rules using a hybrid of ... Expand

Abstract: A system that aggregates case-based linguistic decision rules using a hybrid of the dominance-based rough set approach (DRSA) and the Dempster–Shafer (DS) theory of evidence is proposed for multiple criterion-multiple participant sorting. First, DRSA is employed to infer linguistic decision rules that estimate the preferences of a few participants by means of their evaluations of representative case sets. Next, DS theory is applied to aggregate the decision rules triggered by all participants’ evaluations of an alternative, thereby generating an overall decision recommendation for the alternative. The method is demonstrated on a numerical example. Collapse

Topics: decision making IT work

Methods: computational algorithm

Theories: Dempster-Shafer theory mathematical theory

Formalising Workflow: A CCS-inspired Characterisation of the YAWL Workflow Patterns

2007 | Group Decision and Negotiation | Citations: 0

Authors: Farrell, Andrew D. H.; Sergot, Marek J.; Bartolini, Claudio

Abstract: We present work concerning the formal specification of business processes. It is ... Expand

Abstract: We present work concerning the formal specification of business processes. It is of substantial benefit to be able to pin down the meaning of business processes precisely. This is an end in itself, but we are also concerned to do so in order that we might prove properties about the business processes that are being specified. It is a notable characteristic of most languages for representing business processes that they lack a robust semantics, and a notable characteristic of most commercial Business Process Management products that they have no support for verification of business process models. We define a high-level meta-model, called Liesbet , for representing business processes. The ontological commitments for Liesbet are sourced from the YAWL workflow patterns, which have been defined from studies into the behavioural nature of business processes. A formal characterisation of Liesbet is provided using Milner’s Calculus of Communicating Systems (CCS). In this article, we omit some of the technical details of this characterisation and instead present the essential features by means of an abstract machine language, called LCCS. We also explain how we have facilitated the verification of certain properties of business processes specified in Liesbet , and discuss how Liesbet supports the YAWL workflow patterns. We include a simple three-part example of using Liesbet . Collapse

Topics: knowledge representation business process management machine language extensible markup language web service

Methods: design science metamodel business process modeling ontological modelling computational algorithm

Theories: mathematical theory

Cultural Dimensions and Prototypical Criteria for Multi-Criteria Decision Support in Electronic Markets: A Comparative Analysis of Two Job Markets

2005 | Group Decision and Negotiation | Citations: 0

Authors: Rutkowski, Anne-Francoise; Van De Walle, Bartel

Abstract: Electronic markets are virtual meeting places where buyers and sellers interact ... Expand

Abstract: Electronic markets are virtual meeting places where buyers and sellers interact to trade products or services. The main motivation for both buyers and sellers to participate in an electronic market is the desire to maximize their private utility (Bakos 1998). Electronic markets therefore usually provide some form of communication, decision or negotiation aid for buyers and sellers to support their utility maximizing goals. This paper presents a comparative analysis of two electronic job market case studies conducted at a university in Europe (Brussels, Belgium) and in the United States (Newark, New Jersey). At the occasion of the universities’ career fairs, students (n = 392) and local companies (n = 57) were invited to participate in an electronic job market to identify the best job offers (n = 137) and students, respectively. Participants were able to create personalized software agents to aid their search and decision making activities in the market. Every software agent was embedded with a multi-criteria decision support tool to produce a rank ordered list of students or job offers. Preference data gathered from market participants’ use of the multi-criteria decision model allow us to construct relational preference structures using a technique based on the mathematical theory of fuzzy relations (Bandler and Kohout 1980). These preference structures express relationships among the criteria that students and companies have used to identify job offers and companies, respectively. The purpose of the paper is to present the communicative and cultural implications of these relational preference structures. The theories of Hofstede (1983), Hall (1977) and Trompenaars (1993) on cultural dimensions allow us to discuss cultural differences on the choice of prototypical criteria. The paper concludes with implications for the use of electronic markets in the staffing industry and the role of software agents in such job markets. Collapse

Topics: decision support system electronic market decision making human resource management national culture

Methods: cross sectional research multiple criteria decision making trade-off analysis method mathematical model case study

Theories: fuzzy set theory mathematical theory

Explaining Complex Organizational Dynamics

1999 | Organization Science | Citations: 14

Authors: Dooley, Kevin J.; Van de Ven, Andrew H.

Abstract: Studies of organizational processes can yield observations in the form of event ... Expand

Abstract: Studies of organizational processes can yield observations in the form of event time series that can be analyzed to determine whether they reflect periodic, chaotic, white noise, or pink noise dynamic patterns. These different patterns each imply different underlying generative mechanisms and hence, different process theories. In this paper we present a model that describes how these four dynamical patterns are different from one another. Specifically, a causal system can be characterized by its dimensionality, and by the nature of interaction between causal factors. Low dimensional causal systems yield periodic and chaotic dynamics, while high dimensional causal systems yield white and pink noise dynamics. Periodic and white noise dynamics stem from systems where causal factors act independently, or in a linear fashion, while chaotic and pink noise systems stem from systems where causal factors act interdependently, in a nonlinear fashion. Thus, given a diagnosis of an observed event time series, we can hypothesize a particular story, or causal process theory, that might explain in organization-specific terms why such dynamics came about. In doing so, we also propose that the observation of chaotic organizational dynamics may often signify the presence of control and/or cooperation, rather than a lack of it, as implied by the vernacular use of the term. We conclude by challenging organizational researchers to define new models that capture such observed behavior. Collapse

Topics: innovation management organizational change logistics management

Methods: longitudinal research time series analysis theory development autoregressive moving average model

Theories: mathematical theory

An Outline of the Mathematical Theory of Democracy and Its Applications

1997 | Group Decision and Negotiation | Citations: 0

Authors: TANGUIANE, ANDRANICK

Abstract: This paper summarizes the author‘s studies in mathematical theory ofdemoocracy ( ... Expand

Abstract: This paper summarizes the author‘s studies in mathematical theory ofdemoocracy (Tanguiane 1991, 1993, 1994, 1997). The main consideration is theformalization of the notion of representativeness measured by the weight ofthe coalition represented in each event of decision making. Therepresentativeness is used to estimate the quality of individualrepresentatives (president) and two forms of representative bodies likecabinet (named by analogy with the cabinet of ministers) and council(parliament). In particular, we suggest a solution to Arrow‘s paradox byproving that there always exists an Arrovian dictator who is morerepresentative of the society than a dictator in a proper sense. We alsooutline possible applications of the model to Gallup polls, multicriteriadecision making, and analysis of political situations. Collapse

Topics: decision making individual characteristic IS economics economic surplus

Methods: trade-off analysis method multiple criteria decision making survey descriptive statistic

Theories: mathematical theory