دانلود رایگان مقاله تحلیل ترافیک بصری برای مدیریت شبکه

Abstract

We propose a visual analytic system to augment and enhance decision-making processes of supply chain managers. Several design requirements drive the development of our integrated architecture and lead to three primary capabilities of our system prototype. First, a visual analytic system must integrate various relevant views and perspectives that highlight different structural aspects of a supply network. Second, the system must deliver required information on-demand and update the visual representation via user-initiated interactions. Third, the system must provide both descriptive and predictive analytic functions for managers to gain contingency intelligence. Based on these capabilities we implement an interactive web-based visual analytic system. Our system enables managers to interactively apply visual encodings based on different node and edge attributes to facilitate mental map matching between abstract attributes and visual elements. Grounded in cognitive fit theory, we demonstrate that an interactive visual system that dynamically adjusts visual representations to the decision environment can significantly enhance decision-making processes in a supply network setting. We conduct multi-stage evaluation sessions with prototypical users that collectively confirm the value of our system. Our results indicate a positive reaction to our system. We conclude with implications and future research opportunities.

7. Conclusion

Supply chains are increasingly viewed as complex networks of business relationships, evolving in a bidirectional and nonlinear fashion. The classical logic for SCM based on unidirectional and linear relationships can be significantly crippling in such a complex decision-making context that demands a better understanding of the structural characteristics of supply networks [5]. Accordingly, the emerging network perspective of SCM requires a novel approach for designing DSS and tools for key decision makers—from front line managers to top management—to stay informed about their supply network and business relationships in general. Visual analytics, a new field and a new approach fusing information visualization and data analytics, can boost human cognition for disentangling patterns from a seemingly complex underlying phenomenon [63]. In this paper, we present an integrated system architecture for designing and building an interactive visual analytic DSS for supply network management. Our system architecture includes three key engines: a visual representation engine, an interaction engine, and a descriptive and predictive analytic engine. These three engines have the system user—decision maker or analyst for SCM—in the decision-making loop and empower them to explore the supply network database that accumulates data from daily business operations. We implement a prototype system using a well-established scholarly supply network data source [72] to demonstrate the instantiation of the system architecture. We evaluate the prototype system over three stages that include a conference workshop of methodological experts, an interview with supply chain experts, and a user study with operations and supply chain managers. These evaluation sessions were designed to provide us valuable feedback from visualization experts, SCM experts, and potential target users. The evaluation results indicate positive confirmation regarding the value of the proposed prototype system applied in real-world SCM and operations management contexts.