Evaluating Explanation Function in Railway Crew Rescheduling System by Think-Aloud Test

Yamada, Takeshi; Sato, Tatsuhiro; Tomiyama, Tomoe; Ueki, Nobutaka

doi:10.1109/iiai-aai.2016.93

Cited by 3 publications

(1 citation statement)

References 17 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The work reported here extends our previous works by adding research findings on a revised cognitive model, sample local solution, and corresponding test results. Related work and problems with conventional systems are reviewed in Section 2.…”

Section: Introductionsupporting

confidence: 66%

Interactive system with explanation function for real‐time railway crew rescheduling

Yamada

Sato

Tomiyama

et al. 2017

IEEJ Transactions Elec Engng

Self Cite

View full text Add to dashboard Cite

To win a user's trust while railway crew rescheduling by interactively explaining the reason(s) for the solution calculated by a computer, we propose a schema for describing crew rescheduling patterns in an 'if-then-because' format, an improved method for controlling user interaction based on the patterns, and a cognitive model of the user in a state-transition form. An improved system developed using the proposed method generates crew schedules automatically by using a rule base comprising 30 patterns. Characteristic of the improved system is the use of 'because' information for explaining the computing process. System tests with an actual schedule for crews and trains in a railway line demonstrate that interactive processing with the help of an explanation function better visualizes the computing process. We evaluate our improved system by think-aloud testing. Protocol analysis of verbal comments revealed that the user state was transited in accordance with our design intent that the user state for interpreting 'because' information plays a hub role in the decision making: users spent 28% of their time interpreting 'because' information. Questionnaire survey results show that contribution of the explanation is 34%. Interactive rescheduling with the help of an explanation function is practical for real-time railway crew rescheduling.

show abstract

Section: Introductionsupporting

confidence: 66%

Interactive system with explanation function for real‐time railway crew rescheduling

Yamada

Sato

Tomiyama

et al. 2017

IEEJ Transactions Elec Engng

Self Cite

View full text Add to dashboard Cite

show abstract

Designing a generic human-machine framework for real-time supply chain planning

Gaudreault

Quimper

Marier

et al. 2016

Journal of Computational Design and Engineering

View full text Add to dashboard Cite

Mixed-Initiative-Systems (MIS) are hybrid decision-making systems in which human and machine collaborate in order to produce a solution. This paper described an MIS adapted to business optimization problems. These problems can usually be solved in less than an hour as they show a linear structure. However, this delay is unacceptable for iterative and interactive decision-making contexts where users need to provide their input. Therefore, we propose a system providing the decision-makers with a convex hull of optimal solutions that minimize/maximize the variables of interest. The users can interactively modify the value of a variable and the system is able to recompute a new optimal solution in a few milliseconds. Four real-time reoptimization methods are described and evaluated. We also propose an improvement to this basic scheme in order to allow a user to explore near-optimal solutions as well. Examples showing real case of how we have exploited this framework within interactive decision support software are given. Highlights A Mixed Initiative System adapted to business optimization problems is presented. Real-time reoptimization methods are described and evaluated. The system is able to recompute a new optimal solution in a few milliseconds. Improvement to this basic scheme allow a user to explore near-optimal solutions. Examples showing real case of exploiting this framework are given.

show abstract