Deep Learning of Spatiotemporal Patterns for Urban Mobility Prediction Using Big Data

2022 | Information Systems Research | Citations: 0

Authors: Wang, Yun; Currim, Faiz; Ram, Sudha

Abstract: Timely and accurate prediction of human movement in urban areas offers instructi ... Expand

Abstract: Timely and accurate prediction of human movement in urban areas offers instructive insights into transportation management, public safety, and location-based services, to name a few. Yet, modeling urban mobility is challenging and complex because of the spatiotemporal dynamics of movement behavior and the influence of exogenous factors such as weather, holidays, and local events. In this paper, we use bus transportation as a proxy to mine spatiotemporal travel patterns. We propose a deep-learning-based urban mobility prediction model that collectively forecasts passenger flows between pairs of city regions in an origin-destination (OD) matrix. We first process OD matrices in a convolutional neural network to capture spatial correlations. Intermediate results are reconstructed into three multivariate time series: hourly, daily, and weekly time series. Each time series is aggregated in a long short-term memory (LSTM) network with a novel attention mechanism to guide the aggregation. In addition, our model is context-aware by using contextual embeddings learned from exogenous factors. We dynamically merge results from LSTM components and context embeddings in a late fusion network to make a final prediction. The proposed model is implemented and evaluated using a large-scale transportation data set of more than 200 million bus trips with a suite of Big Data technologies developed for data processing. Through performance comparison, we show that our approach achieves sizable accuracy improvements in urban mobility prediction. Our work has major implications for efficient transportation system design and performance improvement. The proposed deep neural network structure is generally applicable for sequential graph data prediction. Collapse

Topics: big data global navigation satellite system distributed computing Apache Hadoop transportation management

Methods: longitudinal research time series analysis deep learning artificial neural network cluster analysis

The Midlife Crisis of IT: Accurate Versus Inaccurate Perception of IT Effects

2021 | HICSS | Citations: 0

Authors: Bayer, Michael; Haug, Anders; Lombardi, Jolina; Hvam, Lars

Abstract: Information technology (IT) has been established as a major enabler for business ... Expand

Abstract: Information technology (IT) has been established as a major enabler for business performance. However, studies of the effects of IT typically involve the implicit assumption that the effects reported by the companies studied adequately and accurately describe IT’s effects. In this paper, we challenge this perspective by arguing that in many cases the perceived and actual effects of IT are different. Although most IT researchers likely recognize the discrepancy, this topic has not received much attention. Thus, through a case at a world-leading logistics company, we provide evidence that perceived and actual effects of IT can differ. On this basis, we develop a set of models that describe such discrepancies and use theories from the psychological literature to explain why the discrepancies occur. In this context, we use the term “midlife crisis” as a metaphor for what happens to decision makers’ perception of IT systems after a period of use. Collapse

Topics: productivity systems implementation logistics management enterprise resource planning sensemaking

Methods: case study qualitative interview personal interview single case study descriptive statistic

Using a FMEA–TIFIAD Approach to Identify the Risk of Railway Dangerous Goods Transportation System

2021 | Group Decision and Negotiation | Citations: 0

Authors: Huang, Wencheng; Li, Yueyang; Kou, Xingyi; Wang, Wenzhe; Xu, Yifei

Abstract: In this paper a systemic approach that combines Failure Mode and Effect Analysis ... Expand

Abstract: In this paper a systemic approach that combines Failure Mode and Effect Analysis (FMEA) and trapezoidal intuitionistic fuzzy information axiomatic design (TIFIAD) is proposed to identify the risk of railway dangerous goods transportation system (RDNGTS), which has seven steps: (1) identify potential risk factors and risk sub-indicators of RDNGTS; (2) define the failure criterion of severity of failures (S), portability of occurrence (O) and possibility of detection (D) of FMEA based on trapezoidal intuitionistic fuzzy numbers (TrIFNs); (3) invite experts to score the TrIFNs of S, O and D for each risk sub-indicator; (4) aggregate the TrIFNs of S, O and D for each risk sub-indicator; (5) calculate the TIFIAD of each risk sub-indicator; (6) calculate the total TIFIAD of each risk sub-indicator; (7) calculate the total TIFIAD of each risk factor. Compared with FMEA, FMEA–TIFIAD has the following improvements: (1) The TrIFNs are applied to score the S, O and D of each risk sub-indicator of RDNGTS; (2) The TIFIAD of the S, O and D are calculated for each risk sub-indicator, the product of the S, O and D could replace the Risk Priority Number (RPN) of FMEA; (3) The Entropy Weight Method is used to calculate the weight of each risk sub-indicator. The calculation results show that the potential human risk should be paid more attentions; the FMEA–TIFIAD is more reliable and accurate than FMEA. Finally the improvement suggestions for managers and operators of RDNGTS are given. Collapse

Topics: transportation management logistics management

Methods: case study literature study multiple case study

Theories: fuzzy set theory

Service-oriented intelligent group decision support system: Application in transportation management

2014 | Information Systems Frontiers | Citations: 0

Authors: He, Shiwei; Song, Rui; Chaudhry, Sohail S.

Abstract: In today’s ever changing consumer driven market economy, it is imperative for pr ... Expand

Abstract: In today’s ever changing consumer driven market economy, it is imperative for providers to respond expeditiously to the changes demanded by the customer. This phenomenon is no different in the transportation sector in which a service-oriented Group Decision Support System (GDSS) provides an important role in transportation enterprise to effectively manage and rapidly respond to the varying needs of the customer. In this paper, we explore the integration problem of service-oriented system and intelligence technology through the use of a GDSS. Initially, we analyze a service-oriented architecture and then, propose the design architecture of a service-oriented GDSS. Next, we put forward a general framework that integrates the intelligent techniques as a component into the architecture of service oriented GDSS. In addition, we illustrate how Artificial Intelligence techniques can resolve the conflicts of distributed group decisions. The paper is concluded by providing a number of applications in the railway management system that demonstrates the benefits of the utilization of a service oriented intelligent GDSS. Collapse

Topics: group decision support system transportation management artificial intelligence service oriented architecture fuzzy logic

Methods: computational algorithm optimization model stochastic simulation simulation

Theories: fuzzy set theory

A link-focused methodology for evaluating accessibility to emergency services

2014 | Decision Support Systems | Citations: 0

Authors: Novak, David C.; Sullivan, James L.

Abstract: This paper considers the development of a new measure for evaluating accessibili ... Expand

Abstract: This paper considers the development of a new measure for evaluating accessibility to emergency services via a road network called critical closeness accessibility (CCA). CCA quantifies the relative importance of each link in a roadway network with respect to its system-wide contribution to emergency service accessibility and is based on the network science/graph theory concept of closeness. CCA is a link-focused measure that used to evaluate accessibility on a link-by-link basis and can be used for disconnected networks. We introduce a relative importance-based weighting approach for both origins and destinations that can be applied to different types of nodes including generic link endpoints that represent individual intersections and roadway segments in GIS-based roadway maps as well as to physical locations. CAA offers a unique approach for evaluating accessibility that accounts for the spatial distribution of emergency service locations, the topology of the road network, and the engineering characteristics of the road network such as road types, capacities, volumes, and travel speeds. Collapse

Topics: transportation management social network logistics management

Methods: literature study network analysis

Theories: graph theory

Editorial

2011 | International Journal of Information Management | Citations: 0

Authors: Hills, Philip J.

Abstract: EditorialThis issue contains papers covering a range of issues important to inf ... Expand

Abstract: EditorialThis issue contains papers covering a range of issues important to information management. Our first paper is entitled "On the governance of information: introducing a new concept of governance to support the management of information." It is written . In this paper, the authors describe a definition for Information Governance, extending the common, one-dimensional approach into a more generic statement. Starting from the basic principles of IT-governance four hypotheses are proposed.Next, we have a paper concerned with security issues. Entitled "Reinforcing the security of corporate information resources: a critical review of the role of the acceptable use policy" it is written Collapse

Topics: IT career cloud computing Second Life virtual world innovation management

Methods: literature sample case study field study cross sectional research survey