The results of proof of concept using Motivel

Validity verification of Motivel that visualizes motivation in 5 seconds with only voice.

From August to the end of December 2020, Risk Measurement Technologies Co., Ltd. (RimTech) conducted a proof of concept using “Motivel that visualizes motivation in 5 seconds with only voice” through LIP. Yokohama (* 1). We sincerely thank you to everyone who participated in this proof of concept. In this proof of concept, we investigated the relationship between the motivation indicator calculated by Motivel and activity motivation, concentration / attention, and minor incidents.

* 1 “LIP. Yokohama” means the Yokohama Life Innovation Platform, which aims to continuously create innovations in the health and medical fields from Yokohama, for industry, academia, government, and government to work together. Platform. LIP. Yokohama URL: https://www.city.yokohama.lg.jp/business/keizai/lifescience/lip/lifepf.html

Contact
Risk Measurement Technologies Co., Ltd.
URL: https://rimtech.co.jp/
E-mail: info@rimtech.co.jp

Summary

Operational risk and human resource risk have the problem that it is difficult to visualize the motivation that is a risk factor, and it is difficult to establish an effective risk management method.

Therefore, we will verify the effectiveness of Motivel as a visualization tool that visualizes motivation in 5 seconds with only voice, and verify the possibility of developing a new risk management method from the obtained data.

The verification data was collected from the participants collected through “Public offering through LIP. Yokohama” and “Our original recruitment”. As a result, we were able to collect data on 7,350 utterances and 845 questionnaire responses (period: August-December 2020).

Analysis results: It was found that motivation for activity and concentration / attention generally decreased when the motivation indicator was low. In addition, it was found that the risk of decrease in activity motivation and concentration / attention decreases when the average value increases by more than 5 pts. On the other hand, it was found that the risk of hiyari hat is likely to occur when the volatility of the motivation indicator is low and falls below 20 pts for the third consecutive term.

Next step: Based on the results of this proof of concept we will proceed with the development of Motivel to improve convenience and user experience. We will announce when the new Motivel will be released.

Background and purposes

Risk is a factor that reduces the productivity of society and companies. In recent years, the magnitude of the impact of human-derived risks has attracted attention, and many companies and organizations are trying to reduce risks through health management. However, it was extremely difficult to visualize motivation, which is a factor of human-derived risk, frequently and easily, which hindered the development of data-based risk management methods.

Therefore, we have started a demonstration experiment to verify the effectiveness of Motivel as a tool for visualizing motivation frequently and easily, and to verify whether the obtained data is useful for risk management. We will clarify the quantitative relationship between Motivel data and activity motivation, concentration / attention, and minor incidents to break through risk management methods.

Target risk

Operational risk

It means direct or indirect loss (including opportunity loss) due to incidents, mistakes, fraud, etc.

Human resources risk

It means direct or indirect loss (including opportunity loss) due to turnover or leave. It also applies to situations where performance is low.

The results of proof of concept

Participant

140 people participated through recruitment through LIP. Yokohama (* 2) and our own recruitment. In the questionnaire, we received answers about activity motivation, concentration / attention, and minor incident.

* 2 “Easy measurement of mental health with just voice” We are looking for companies to cooperate in the proof of concept. (https://rimtech.co.jp/index.php/2020/07/21/blog031/)

Data analysis

Dataset for analysis

Throughout the demonstration experiment period, we were able to obtain 7350 records and 845 questionnaire responses. As a result of data scrutiny, we adopted the data collected in October-December (3642 records) for the data set for analysis.

Number of measurement (Unit: Number of cases)

Multivariate analysis

Based on a univariate analysis conducted separately, we performed a multivariate analysis by combining indicators that are effective. We used logistic regression.

Motivation

Correlation can be confirmed in the “level” and “change in average contrast” of the motivation indicator.

Concentration / attention

Correlation can be confirmed in the “level” and “change in average contrast” of the motivation indicator.

Minor incidents

Correlation can be confirmed in the volatility of the motivation indicator. However, since the relationship is weaker than the motivation for activity, concentration, and attention, further analysis is considered necessary.

Below 20 pts for 3: Below 20 points for 3 consecutive terms

Data distribution (box plot)

Motivation (1 is a sample with reduced motivation for activity)

Distribution of mean values of motivation indicator

Concentration / Attention (1 is a sample with reduced concentration and attention)

Distribution of mean values of motivation indicator

Minor incidents (1 is a sample with minor incident)

Distribution of motivation index volatility

Consideration

Based on the results of this proof of concept, the relationship between the motivation indicator, activity motivation, concentration / attention, and hiyari hat can be considered as follows.

Next step

Based on the results of this proof of concept, we will proceed with the development of Motivel to improve convenience and user experience. We will announce when the new Motivel will be released.

(Reference)

Indicators

Motivel calculates four indicators: motivation, vitality, relaxation, and pleasure.

Motivation

The higher the value, the more motivated you are to work, and the higher your attention and concentration. The value of vitality is high, and the larger the vertical fluctuation, the higher the value.

Vitality

The higher the value, the more energetic (comfortable). In addition, the more energetic the state, the more the numerical value moves up and down (the sensitivity of the mind is high).

Relaxation

The higher the value, the calmer the state. In addition, the more energetic the state, the more the numerical value moves up and down (the sensitivity of the mind is high).

Pleasure

The higher the value, the happier you are. In addition, the more energetic the state, the more the numerical value moves up and down (the sensitivity of the mind is high).

Mechanism

Relationship between each indicator

For motivation, the higher the level of vitality and the higher the volatility (fluctuation of indicators), the higher the value.

Vitality is calculated from relaxation and pleasure. The higher the relaxation and pleasure value, the higher the vitality value. The more energetic people are, the greater the variation in measured values tends to be.

Motivel data (as of the end of December 2020)

There are 18744 records. The unit is the number of cases.

Measurement duration

The duration is within 5 seconds in 90% of cases. The unit is millisecond.

[End]

Follow me!