Noriyuki Kushiro

A Method for Generating Ontologies in Requirements Domain for Searching Data Sets in Marketplace

Methods for helping the stakeholders choose appropriate data sets, are indispensable for incubating the market of data into the real world. In this paper, we propose methods for searching appropriate data sets in the marketplace by utilizing both an ontology generated from requirements of the stakeholders and a structured interview with denial inquiries on the ontology.

A Long-term Data Collection System for Life Pattern Sensor

Abstract This paper describes a long-term filed data collection system from a person living house required for designing a life pattern sensor. The life pattern sensor is an activity recognition sensor both for measuring electric power consumption and detecting use of electronic devices by utilizing high-frequency electrical current waveform as a signature of each device. A prototype of the life pattern sensor has been developed and evaluated in five homes. The results show the life pattern sensor measured power consumption within 1% error and identified uses of devices with over 95% accuracy. The capability of the sensor is enough for realizing home energy management system. The long-term data collection system is constructed so as to collect annotated field data for sophisticating the sensor in practical level. In this paper, the overview of the data collecting system and discussion about the following questions based on the field data: 1. Can current waveform be in stable at any conditions in a house? 2. Can sequence of electronic devices describe residents’ real life pattern? 3. Can residents manage energy efficiently by knowing their own life pattern?

A System Design Method Based on Interaction between Logic and Data Sets

A design of cyber physical system is one of the promising domains, where the market of data is expected to penetrate. However, there are risks that any design hypotheses might be supported with data sets in the market, when engineers focus on a particular part of the data intentionally or accidentally. Because, the data sets, especially big data, include diverse kinds of variables to mislead reasoning of the design hypotheses. The validity could not be estimated by the data themselves but could be confirmed only by robustness of the logic behind the design argumentation. In this paper, we propose a system design method with data sets in the market and apply the proposed method to design process for a diagnosis system for semiconductor equipment.

Model Oriented System Design on Big-data☆

Abstract There are risks that any design hypotheses could be supported with big-data, when engineers focus on a particular part of the data intentionally or accidentally, for the reason that big-data include huge and various kinds of data to mislead the reasoning of the hypotheses. The design process of diagnosis system for vacuum pumps in semiconductor factories is picked up as a target of case study. Errors of the hypotheses in the design are clarified by visualizing reasoning process of the design. The visualization of the reasoning process guides the engineer to elaborate the proper design models on the correct hypothesis through cycles of deductive and inductive reasoning with both data and their domain knowledge. The diagnosis system is re-designed and implemented on the established design models and the accuracy of diagnosis of the system is confirmed through the field test. We emphasize that the design method led by the design model on the deep domain knowledge is indispensable for designing system on big-data in the paper.

Experimental Evaluation of a Fast Demand Response System for Small/Medium-Scale Office Buildings

Since the Great East Japan Earthquake, We have acknowledged the need for demand response. Therefore, we developed a system of experimental evaluation of fast demand response (fast DR) for small-scale buildings. Using this system, we confirmed that it is possible to achieve more than 10% reduction in energy consumption by controlling the air conditioning. In this paper, we describe outline of the fast DR system and discuss the result of evaluation.

Aware Environment for Workshop with Game Storming

A context aware environment for workshop with game storming was developed with vision sensors and was evaluated its capability for supporting facilitator of workshop through feasibility test. In this paper, overview of the aware environment for workshop, basic capability of the environment, and results of feasibility test are described.

A Basic Study for Realizing Life Event Sensor for Home Energy Management System

This chapter describes a long term field data collection system from a person occupied house required for designing a life event sensor. The life event sensor is an activity recognition sensor both for measuring electric power consumption and detecting uses of electronic devices by monitoring high frequency electrical waveform emitted by the devices. A prototype of the life event sensor has been developed and evaluated in five homes prior to this study. The results show the sensor measured power consumption within 1% error and identified uses of devices with over 95% accuracy. The basic capability of the sensor is enough for realizing a home energy management system. The long term field data collection system is constructed so as to collect the annotated field data to enhance the sensor for commercial release. The system consists of three sub-systems and collects parameters of power line, status of electronic devices and life events of residents continually. The system has been installed into two houses and the field test has been started for collecting the annotated field data for one year. An overview of the long term field data collection system and discussion about issues for realizing the life event sensor are described in this chapter.

Tool for Extracting Latent Field Overseers’ Knowledge for Risk Recognition on Eyes and Algorithm for Structuring Meta and Domain Knowledge from Utterances

Abstract We have developed a tool for extracting risk field overseers’ recognition knowledge with an eye tracking sensor and a motion sensor, and an algorithm for structuring knowledge from their utterances as propositional network. We applied the tool on risk recognition training for 12 field overseers, including experts and novices, in a construction company and collected data for their eyes and utterances during the training. Typical experts’ process for perceiving risk scenario from each scene (Meta knowledge) and individual pieces of risk knowledge including each risk scenario (Domain knowledge) were elicited on the data with the algorithm. As the results of these, we confirmed that the eyes promote verbalizing field overseers’ knowledge for risk recognition and the propositional network helps structuring their domain and meta knowledge for risk recognition.

Video based risk recognition training tool using eye tracking device

Risk recognition training is an important training in construction companies to avoid work-related accidents. The existing risk recognition training is usually conducted on an illustration including explicit risks. However, it is difficult to express dynamic scenarios and surroundings of the scene due to the still illustration. In this study, a video based risk recognition training tool with an eye tracking device has been developed. We applied the tool to discover differences of attentions between veterans and we found out that veterans to do risk recognition based on meta-knowledge (risk recognition process) and domain-knowledge (individual knowledge of construction work).

Initial Practice of Telematics-Based Prognostics for Commercial Vehicles: Analysis Tool for Building Faults Progress Model for Trucks on Telematics Data

Abstract Unscheduled maintenance of commercial vehicle brings users increasing maintenance cost, decreasing reliability of their fleet and quality of customer service. The prognostics systems are required for avoiding unexpected breakdowns in the vehicles. As an initial practice for realizing the prognostics system, a fault progress model to unexpected breakdowns was built on correlations among cautions/warnings from sensors within telematics data in this study. An analysis tool was developed for constructing the model, and the model extracted from the tool was evaluated its validity through reviews by engineers in a vehicle manufacturer with different specialties.