JPS60142837A - Working mistake generation forecasting method and apparatus - Google Patents

Abstract

(57) [Summary] This bulletin contains application data before electronic filing, so abstract data is not recorded.

Description

[Detailed Description of the Invention] (a) Abstraf 1 - This invention relates to a method and device for predicting the occurrence of work errors based on pulse rate, and its characteristics include continuous measurement of pulse rate intervals. By calculating the amount of time that indicates the degree of variation in the pulse interval time at predetermined intervals, and by setting up an alarm when a sudden change in this amount is detected, it is possible to simply Prevent increases in work errors by simply measuring continuously 111
There is a point C where it is possible to do 1.

(0) Industrial application fields: Typists, Keypange V-1 assemblers, inspection engineers, car drivers, and other workers can predict the increase in work errors just in advance and warn them or remind them to take a break. By taking appropriate measures to prevent work errors, such as this, it is possible to naturally prevent an increase in work mistakes.

(c) Technical background It has been known that there is a close relationship between the occurrence of work errors and fluctuations in pulse rate. , changes in emotions, and other factors, and also fluctuate slowly over time, making it impossible to grasp exactly what this relationship is.

As a result of repeated experiments focusing on the variation in pulse rate, the present inventor found that after 5 minutes to 15 minutes after the variation in pulse rate suddenly changed.
We discovered that the occurrence of work errors increases rapidly after about a minute.

(d) Purpose of the Invention The object of the present invention is to provide a method and device for predicting the occurrence of work errors that utilizes the fact that variations in pulse rate change rapidly prior to an increase in work mistakes.

(e) Embodiment of the method An embodiment of the method of the present invention will be described with reference to FIGS. 1 and 2.

The pulse rate of the worker during work is continuously detected, the pulse rate interval time tI, [2, L3... is measured, and each pulse rate I! Pulse rate P1-60/l + for every g interval,
P2 = 60/l 2 , P3 = 60/13, etc. are calculated.

Next, the average value Q='(P+ +P2 +-+P11)/n of this pulse rate for the past minute is calculated every minute, and is used as standard basic data of the pulse rate for that one minute.

The standard deviation S of the pulse rate per minute is a quantity that indicates the degree of variation in the pulse rate, and work errors suddenly occur about 5 to 15 minutes after a sudden change in the standard width difference is observed. Experimental results have shown that it increases.

Therefore, in order to monitor the fluctuation of the standard shoulder difference S of the pulse rate for one minute, the following manual measurement is taken.

The standard deviation of the pulse rate per minute may increase temporarily (e.g., 1 to 2 minutes) even when the worker is surprised by something; it is important to predict such work errors. In order to remove the influence of non-essential temporary changes in the standard deviation, the data are smoothed by taking a 7-point moving average of the standard deviation. That is, 100 + = (S + 10S 2+...10S7)/7002=(S2+S3+...+s a )/700□ -(
s-+s +1 +...+81±6)77 Next, the second
As shown in (a) in the figure, based on 40 consecutive smoothed standard deviations, the average is 1 as follows:
fiX + +13 and standard deviation σ1 are calculated.

Then, X1±3σ1 is taken as a comparison reference value, and the next smoothed standard B difference 341 is compared with this comparison reference value, and if the former is within the latter range, it is determined that there is no sudden change.

Furthermore, when one minute has passed, as shown in (b) in the figure, a new smoothed target 1$ deviation 341 is adopted, and the first smoothed target IP=deviation 91 is discarded. , create a new average 11αX2 and standard deviation σ2, and compare the next smoothed standard deviation S4z with the reference value X2
Compare ±S−σ2.

Repeat this operation one after another every minute, and if the smoothed standard L% m m s goes out of the range of the comparison standard value X ± 3σ, then the smoothed standard deviation 3 suddenly changes. Make a judgment and issue a warning.

The reason why the average value X 83 and standard deviation σ are updated every minute is to remove gradual fluctuations in the smoothed standard deviation. Even if X1±3σ1, which is set to
In practical terms, a satisfactory result can be obtained.

Also, based on 40 smoothed standard width differences,
3 and σ, but it is smaller than this (for example, 3
It may be calculated based on 0) data g@.

J3, instead of adopting X±3σ as the reference value for comparison,
X±2σ, X±σ, etc. may be adopted, and may be determined as appropriate in consideration of the type of work and other factors.

(F) Embodiment of the Apparatus FIG. 3 shows an embodiment of the apparatus of the present invention being used by a typist at work.

The work error occurrence prediction device includes a pulse rate detecting sensor 1, a pulse rate measuring device 2, a microcomputer 3, and an alarm device 4.

Pulse detection sensor 1 using changes in infrared transmittance
is attached to the earlobe of the typist 5, as shown in FIG.

In the pulse rate measuring device 2, from this signal, the pulse rate interval t,
, t2. ... is measured and this is sent to the microcomputer 3.
send to

The microcomputer 3 calculates the pulse rate P, the average value Q and standard deviation S of the pulse rate per minute, and calculates the standard deviation g+, g2. ... is calculated one after another, and after 46 minutes have elapsed, the first comparison reference value X1i3σ1 is calculated, and then the calculated average ? The visualized standard deviation 341 is compared with this comparison reference value×1±3σ1.

Then, every minute after that, a new comparison base Hz H direct
Calculate ±3σK and the next smoothed target! 1 (deviation E
The operation of comparing <++o with this new comparison reference value is repeated, and when the former is out of the range of the latter, an operation signal is sent to the alarm device 4 (by known means such as a blinking display or a buzzer) and an alarm is emitted.

The pulse rate detection sensor 1 may be placed at any suitable location within the earlobe, and the sensor 1 may be placed at any suitable location in the song. Can be used.

In the above example, the standard deviation S of the velocity of pulse 1 per minute is the average value. Although a 7-point moving average is used for visualization, other well-known methods can be used as appropriate to smooth the data, and even if smoothing is not performed, , equivalent results can be obtained.

In addition, standard deviation is used as a quantity that indicates the degree of variation in pulse rate, but it is not limited to this. For example, the average deviation of pulse rate (the average of the absolute value of the difference between pulse rate and its average) ).

Furthermore, instead of using the velocity of [pulse] as the basic data for analysis,
Similar results can be obtained by using the pulse interval time as the basic data.

(1-) Effects As explained above, the method and device for predicting work errors of the present invention can prevent an increase in work mistakes from occurring with high accuracy by simply continuously measuring the worker's pulse. It can be predicted and has a great effect on preventing accidents caused by work errors.

[Brief explanation of the drawing]

Figure 1: Graph explaining pulse interval time Figure 2:
Figure 3 shows a graph showing the smoothed standard deviation...Block diagram of an industrial error prediction device according to an embodiment of the present inventionFigure 4...Enlarged diagram 1・
...Pulse detection sensor 2...Pulse measuring device 3...
Microcomputer 1... Alarm 5... To Tivis 1 Agent Patent attorney Makoto Oka - Figure 1 Figure 3 Figure 4

Claims (3)

[Claims] (1) The feature is that the pulse interval time is continuously measured, a quantity indicating the degree of variation in the pulse interval time is determined at predetermined intervals, and an alarm is issued when a sudden change in the quantity is detected. Payment for advance notice of work errors. (2) The method for predicting the occurrence of work errors as set forth in claim 1, wherein the degree of variation in pulse interval time is determined by the standard deviation of pulse rate traces. (3) A pulse rate detection sensor that detects the worker's pulse rate, a pulse rate measuring device that measures the pulse interval time based on the signal sent from the pulse rate detection sensor, and a pulse rate measurement device that measures the pulse rate interval time every predetermined interval. It consists of a microcomputer that calculates tffi, which indicates the degree of This is a device for predicting the occurrence of work errors.