JP2007193575A - Device for evaluation of measure-against-risk - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To realize effectiveness in disaster countermeasure expenses by considering a user's tendency such as avoiding risks, without troublesomeness of a user's inputting. <P>SOLUTION: A device for evaluation of measure-against-risk is provided with a computation means 214 which selects one piece of information from information on a plurality of phenomena stored in an amount-of-damage/degree-of-improvement DB and one piece of information from information obtained after the measures against the phenomena are taken, generates query information on payable amounts to users for the phenomena to output it based on the selected information, estimates convenience-and-benefit functions of the users based on answer information to the query information input from the users, and determines whether ending conditions are satisfied based on information on the results of the estimations. Further, if the ending conditions are not satisfied, the computation means selects one piece of information again from the information on the plurality of phenomena and one piece of information from the information obtained after the measures against the phenomena are taken. If the ending conditions are satisfied, based on the estimated convenience-and-benefit functions the effectiveness to anti-risk expenses for the selected phenomena is computed to be output. <P>COPYRIGHT: (C)2007,JPO&INPIT

Description

  The present invention relates to a risk countermeasure evaluation apparatus for evaluating cost-effectiveness of countermeasures for preventing risks due to disasters such as earthquakes and fires, terrorism, etc., and in particular, it is a technology effective when applied to reduction of user input. It is about.

  Conventionally, as a method for evaluating risk countermeasures, for example, there is a system described in JP-A-2004-280444 (Patent Document 1). This method evaluates the countermeasure by comparing the effect of reducing the risk and the cost required for the countermeasure.

  The outline of the conventional evaluation method is shown below.

If the occurrence probability of event i is pi, the estimated damage amount is Xi, and the estimated damage amount of event i by countermeasure j is Yij, the predicted damage improvement amount is calculated as the total amount Σpi × (Xi−Yij) for event i. Is compared with the cost Cj for the measure j.

As a comparison method, a simple comparison (Σpi × (Xi−Yij)) − Cj, ROI (
(Σpi × (Xi−Yij)) − Cj) / Cj.

On the other hand, in recent years, there is a lot of debate about whether it is desirable to use a cost-benefit method using a benefit function instead of evaluating the risk countermeasures as described above for the risk countermeasures. Katsumi, “A new flood risk assessment and cost-benefit analysis in flood control economic research”, JICE REPORT 2002 / No. 2, National Institute of Land and Technology (Non-patent Document 1) are discussed.
JP 2004-280444 A Katsumi Yukawa, “A new flood risk assessment and cost-benefit analysis in flood control economic surveys”, JICE REPORT 2002/2, National Institute for Land and Infrastructure Research

  However, in the risk countermeasure evaluation using the conventional benefit function, it is necessary to input data from the user for various events in order to reflect the user's benefit. It took time and effort.

  Therefore, an object of the present invention is to provide a risk countermeasure evaluation apparatus that can reflect the benefits of the user in risk countermeasure evaluation and can reduce the input of the user at that time.

  Of the inventions disclosed in the present application, the outline of typical ones will be briefly described as follows.

  The risk countermeasure evaluation apparatus according to the present invention includes, for each of a plurality of events, a database that stores information on the probability of occurrence and expected damage amount, and information on the expected damage amount after one or more countermeasures. Select one information from the stored information of multiple events and one information from the information after countermeasures for that event, and generate question information about the amount payable for the event to the user based on the selected information The user's benefit function is estimated based on the answer information to the question information input by the user, and it is determined whether the termination condition is satisfied based on the information of the estimation result. If there is not, select one information again from the information of multiple events and information after countermeasures for that event, and if it is satisfied, based on the estimated benefit function It is obtained by an arithmetic means which calculates and outputs the cost-effectiveness of the risk measure for the selected event.

  In addition, the risk countermeasure evaluation apparatus according to the present invention includes, for each of a plurality of events, a database that stores information on the probability of occurrence and expected damage amount, and information on the expected damage amount after one or more countermeasures; Question information about the amount payable for an event to a user based on the selected information by selecting one information from information on multiple events stored in the database and information after countermeasures for that event Based on the answer information input from the user, the input / output means for inputting the answer information from the user to the question information, Based on the parameter estimation means for estimating the user's benefit function and the information on the result of estimation by the parameter estimation means, it is determined whether or not the termination condition is satisfied. If the answer is yes, the question generation means again selects one information from the information of the plurality of events and the information after the countermeasures for the event, and if satisfied, the selection is made based on the estimated benefit function And a calculating means for calculating and outputting the cost effectiveness of the risk countermeasure for the event.

  Among the inventions disclosed in the present application, effects obtained by typical ones will be briefly described as follows.

  According to the present invention, it is possible to calculate a cost-effectiveness reflecting a user's preference such as risk avoidance regarding risk countermeasures. Furthermore, by estimating the user's benefit function, it is possible to save the user's input effort.

  Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings. Note that components having the same function are denoted by the same reference symbols throughout the drawings for describing the embodiment, and the repetitive description thereof will be omitted.

<Configuration of risk countermeasure evaluation device>
The configuration of the risk countermeasure evaluation apparatus according to an embodiment of the present invention will be described with reference to FIGS. FIG. 1 is a configuration diagram showing a configuration of a risk countermeasure evaluation apparatus according to an embodiment of the present invention, and FIG. 2 is a hardware configuration diagram when the risk countermeasure evaluation apparatus according to an embodiment of the present invention is implemented by a computer. FIG. 3 is a diagram showing the contents of the damage amount / improvement degree DB of the risk countermeasure evaluation apparatus according to the embodiment of the present invention.

  In FIG. 1, the risk countermeasure evaluation apparatus 110 includes a WTP input / output unit 111, a damage amount / improvement degree DB 112, a parameter estimation unit 113, a damage amount calculation unit 114, a question generation unit 115, and a result output unit 116.

  The WTP input / output unit 111 displays conditions such as the probability of occurrence of risk to the user, and inputs an amount of money (WTP: Wish To Pay) that can be paid based on the displayed condition from the user.

  As shown in FIG. 3, the damage amount / improvement degree DB 112 stores an occurrence probability 520, an expected damage amount 530, damage amounts 540 and 550 when countermeasure j is applied to each event 510.

  The parameter estimation unit 113 estimates a benefit function based on information input from the WTP input / output unit 111 or the like.

  The damage amount calculation means 114 calculates the cost effectiveness of the risk countermeasure from the information input from the WTP input / output means 111 and the information estimated by the parameter estimation means 113.

  The question generation unit 115 generates a question for the user from the information stored in the damage amount / improvement degree DB 112.

  The result output means 116 outputs the result of cost effectiveness calculated by the damage amount calculation means 114.

  Further, the hardware configuration when each configuration of the risk countermeasure evaluation apparatus 110 shown in FIG. 1 is implemented by a computer includes an output unit 211, a data storage unit 212, an input unit 213, and a calculation unit 214 as shown in FIG. Have. Here, a display is generally used as the output unit 211, but an audio output device, a printer, and the like are also conceivable. As the data storage means 212, a volatile memory, a non-volatile memory, etc. can be considered in addition to a hard disk. The input unit 213 may be a keyboard, a voice input device, a pointing device, and the like, and the arithmetic unit 214 may be a CPU, MPU, or the like.

  The WTP input / output unit 111 of FIG. 1 is configured by an output unit 211 and an input unit 213, the damage amount / improvement degree DB 112 is configured by a data storage unit 212, and the result output unit 116 is configured by the output unit 211 and the like. Is done. In addition, the parameter estimation unit 113, the damage amount calculation unit 114, and the question generation unit 115 perform each process when a program or the like is executed by the calculation unit 214.

<Operation of risk countermeasure evaluation device>
Next, the operation of the risk countermeasure evaluation apparatus according to the embodiment of the present invention will be described with reference to FIGS. 4 is a flowchart showing the operation of the risk countermeasure evaluation apparatus according to the embodiment of the present invention, FIG. 5 is a diagram showing an example of the question screen of the risk countermeasure evaluation apparatus according to the embodiment of the present invention, and FIG. It is a figure which shows an example of the result display screen of the risk countermeasure evaluation apparatus which concerns on one embodiment of this invention.

  First, before the calculation according to the flowchart shown in FIG. 4, the occurrence probability pi, the damage amount Xi, and the damage amount Yij when the countermeasure j is applied to the event i by the method described in Patent Document 1, etc., and the cost for the countermeasure j It is assumed that Cj and the like are calculated and stored in the damage amount / improvement degree DB 112.

  First, in step 310, event i and countermeasure j are selected.

  The selection methods of i and j are selected in order, randomly selected, ["question selection for prediction and model selection", Gerhard PAASS, Information Processing Vol. 38, No. 7, pp562-568, July 1997] Various selections such as those described in Non-Patent Document 2) are conceivable, but when the number of events and countermeasures is small, they are selected in order or at random for simplification of processing. If the method is taken and the number of events and countermeasures is large, it is considered that the overall efficiency is improved by selective extraction.

  In step 311, pi, Xi, Yij corresponding to the event i and countermeasure j selected in step 310 are retrieved from the damage amount / improvement degree DB 112, and in step 320, a question to the user is generated by the question generation means 115. The question screen 400 as shown in FIG. 5 is displayed from the WTP input / output unit 111.

  In the example shown in FIG. 5, in addition to the occurrence probability pi of event i, the predicted damage amount Xi, and the post-measurement damage amount Yij, the product of the occurrence probability pi and the predicted damage improvement amount Xi−Yij It is displayed for the convenience of the user, but this is not necessary. Further, in FIG. 5, the desired payment amount is directly asked, but a question such as “Do you want to pay this amount?” Can be considered as in the two-step selective method in the virtual market method.

  The risk countermeasure evaluation apparatus 110 receives the desired payment amount WTPij, which is manually input based on the data displayed in step 320, in step 330.

  Based on the WTPij received in step 330, the parameter of the benefit function is updated by the parameter estimation means 113 in step 340. A widely known parameter estimation method is used for this parameter update.

  In parameter estimation, parameter values are determined so as to optimize predetermined error functions and likelihood functions, and some of them are also introduced in Non-Patent Document 2.

  Parameter estimation includes sequential estimation in which estimation is performed using only the current parameter value and the latest input information, in addition to normal estimation in which parameters are recalculated using all information input so far. Which is used may be determined from the processing performance of the risk countermeasure evaluation apparatus 110, the storable data capacity, and the like.

  In step 350, an end determination is made based on the parameters updated in step 340. If the end condition is not satisfied, the process returns to the selection of event i and countermeasure j in step 310, and if the end condition is satisfied. In step 360, cost-benefit calculation is performed by the damage amount calculation means 114 using the benefit function estimated in step 340. In step 370, the result output means 116 displays a result display screen as shown in FIG. 600 is output and the process ends.

<Specific example of operation of risk countermeasure evaluation device>
Next, a specific example of the operation of the risk countermeasure evaluation apparatus according to the flowchart shown in FIG. 4 will be described based on examples of a benefit function, an error function, a question selection method, and an end condition.

  First, as a benefit function, f (p, X, Y; α) = p (X−Y) + α.

  This is a model that humans may pay slightly more by α than the expected value p (X−Y) for damage reduction.

  In this embodiment, such a benefit function is used to simplify the explanation, but it is possible to estimate the benefit with higher accuracy if α itself uses a complex function that depends on p, X, and Y.

  As the error function, the least square method is used, and is as follows with respect to the number of data k.

1 / kΣ (f (pi, Xi, Yij; α) −WTPij) ^ 2 = [Σ (p (Xi−Yij) +
α-WTPij) ^ 2] / k
The question selection method is random selection, and the end condition is “the absolute value of the amount of change in the estimated value of parameter α is 10000 yen or less or the number of repetitions k exceeds 10.”

  Based on the above specific formula, the specific calculation contents of the flowchart shown in FIG. 4 will be described.

  First, at step 310, i and j are selected at random.

  Here, if i = 1 and j = 1 are selected, event seismic intensity 7 corresponding to i, j selected in step 311, pi = 1%, Xi = 5 million yen, Yij = 4 million yen (541) Extracted from the damage amount / improvement degree DB 112, a question screen is created in step 320, and a question screen 400 as shown in FIG. 5 is displayed.

  If the user inputs that it is possible to pay 1.13 million yen for this countermeasure, the risk countermeasure evaluation apparatus 110 that receives this input in step 330 sets α = 1.1330,000−pi × (Xi as α that minimizes the error function. -Yij) = 130,000.

  In step 350, the end determination is made, but this time is the first time, the previous parameter α does not exist, and since the number of repetitions k = 1, the end condition is not satisfied, and the process returns to step 310.

  Next, at step 310, i and j are selected at random.

  Here, if i = 2 and j = 2 are selected, event seismic intensity 6 strong, pi = 2%, Xi = 2 million yen, Yij = 500,000 yen (542) corresponding to i, j selected in step 311 ) Is extracted from the damage amount / improvement degree DB 112, and a question screen is created and displayed in step 320.

  If the user inputs that it is possible to pay 3,050,000 yen for this countermeasure, the risk countermeasure evaluation apparatus 110 that has received this input in step 330 sets α = 1 / 2Σ (WTPij−pi × X) as α that minimizes the error function. (Xi−Yij)) = 1/2 [(1.13 million-1 million) + (305 million-3 million)] = 90,000.

  In step 350, the end determination is performed. Since the previous estimated value of the parameter α is 130,000 and this time is 90,000, the amount of change is 40,000, which does not satisfy the end condition of 10,000 or less, and the number of repetitions k = 2. The process returns to step 310 without satisfying the termination condition.

  Next, at step 310, i and j are selected at random.

  Here, if i = 3 and j = 1 are selected, event seismic intensity 6 weak, pi = 4%, Xi = 1,000,000 yen, Yij = 500,000 yen (543) corresponding to i and j selected in step 311 ) Is extracted from the damage amount / improvement degree DB 112, and a question screen is created and displayed in step 320.

  If the user inputs that it is possible to pay 2.12 million yen for this countermeasure, the risk countermeasure evaluation apparatus 110 that received this input in step 330 sets α = 1 / 3Σ (WTPij−pi × X) as α that minimizes the error function. (Xi−Yij)) = 1/3 [(1.13 million-1 million) + (305 million-3 million) + (2.12 million-2 million)] = 100,000.

  In step 350, an end determination is made, and since the amount of change in the estimated value of α, which is the end condition, satisfies 10,000 or less, the cost effect is calculated using the calculated parameter α in step 360.

Here, assuming that ROI is used as cost-effectiveness, ROIj = ((Σf (
pi, Xi, Yij; α))-Cj) / Cj. This is calculated for all the countermeasures, and is output as a cost effect on the result display screen 600 as shown in FIG.

  As described above, in the present embodiment, the risk such as the probability of occurrence of risk is displayed to the user, and the amount of money that can be paid based on the displayed condition from the user is input. It is possible to calculate the cost-effectiveness reflecting the user's preference such as risk avoidance regarding countermeasures. Further, by estimating the utility function of the user, it is possible to save the user's input effort.

  As mentioned above, the invention made by the present inventor has been specifically described based on the embodiment. However, the present invention is not limited to the embodiment, and various modifications can be made without departing from the scope of the invention. Needless to say.

  For example, in the present embodiment, the event i has been described as a discrete finite number, but the same can be considered for continuous events such as the magnitude of an earthquake. In this case, the occurrence probability density function p (i) is used instead of the occurrence probability pi, and the part for which the sum is calculated for i is replaced with the integral for i.

  In this embodiment, the event i and the countermeasure j are selected in step 310 and the occurrence probability and the expected damage amount are searched from the damage amount / improvement degree DB 112. However, these are not necessarily required, and are used for the question. The probability or the like may be determined directly.

  In other words, if there are only two possible events and the occurrence probabilities are 1% and 10%, respectively, in this embodiment, only a question about 1% event or a question about 10% event occurs. When the parameters to be used are directly determined, for example, a question about an event having an occurrence probability of 5% can be considered.

  This direct determination method is particularly effective when the question selection described in Non-Patent Document 2 is used when creating question data.

  The present invention relates to a risk countermeasure evaluation apparatus for evaluating cost effectiveness of measures for preventing risks due to disasters such as earthquakes and fires, terrorism, etc., and in particular, the cost effectiveness of risk measures taking human benefits into consideration. It is applicable to a device that performs comparison.

It is a block diagram which shows the structure of the risk countermeasure evaluation apparatus which concerns on one embodiment of this invention. It is a hardware block diagram at the time of mounting the risk countermeasure evaluation apparatus which concerns on one embodiment of this invention with a computer. It is a figure which shows the content of damage amount and improvement degree DB of the risk countermeasure evaluation apparatus which concerns on one embodiment of this invention. It is a flowchart which shows operation | movement of the risk countermeasure evaluation apparatus which concerns on one embodiment of this invention. It is a figure which shows an example of the question screen of the risk countermeasure evaluation apparatus which concerns on one embodiment of this invention. It is a figure which shows an example of the result display screen of the risk countermeasure evaluation apparatus which concerns on one embodiment of this invention.

Explanation of symbols

  DESCRIPTION OF SYMBOLS 110 ... Risk countermeasure evaluation apparatus, 111 ... WTP input / output means, 112 ... Damage amount / improvement degree DB, 113 ... Parameter estimation means, 114 ... Damage amount calculation means, 115 ... Question generation means, 116 ... Result output means, 211 ... Output means, 212... Data storage means, 213... Input means, 214.

Claims (5)

For each of a plurality of events, a database that stores information on the probability of occurrence and the expected damage amount, and information on the expected damage amount after one or more countermeasures,
One information is selected from the information of the plurality of events stored in the database and information after countermeasures for the event, and payment of the event to the user is possible based on the selected information Generate and output question information about the amount, estimate the benefit function of the user based on the answer information to the question information input from the user, and set the end condition based on the information of the estimation result If it is not satisfied, if it is not satisfied, one information is selected again from the information of the plurality of events and information after countermeasures for the event, and if it is satisfied, the estimated benefit A risk countermeasure evaluation apparatus comprising: a calculation means for calculating and outputting a cost effectiveness of the risk countermeasure for the selected event based on a function. For each of a plurality of events, a database that stores information on the probability of occurrence and the expected damage amount, and information on the expected damage amount after one or more countermeasures,
One information is selected from the information of the plurality of events stored in the database and information after countermeasures for the event, and payment of the event to the user is possible based on the selected information Question generating means for generating question information about the amount;
Output the question information by the question generating means to the user, input / output means for inputting the answer information from the user to the question information,
Parameter estimation means for estimating a benefit function of the user based on the answer information input from the user;
It is determined whether an end condition is satisfied based on information of a result of estimation by the parameter estimation unit, and if not satisfied, one information from the plurality of event information is re-executed by the question generation unit and a countermeasure for the event And calculating means for calculating and outputting the cost-effectiveness of the risk countermeasure for the selected event based on the estimated benefit function when one information is selected from the later information and is satisfied. Risk measure evaluation device characterized by In the risk countermeasure evaluation apparatus according to claim 1 or 2,
The risk condition evaluation apparatus, wherein the termination condition is information on the number of repetitions and a parameter change amount in the estimation of the benefit function. In the risk countermeasure evaluation apparatus according to claim 1, 2, or 3,
A risk countermeasure evaluation apparatus, wherein when selecting one information from information of a plurality of events and one information from information after countermeasure of the event, selection is performed using order, random, or selective extraction. In the risk countermeasure evaluation apparatus according to claim 1, 2, 3, or 4,
The risk countermeasure evaluation apparatus characterized in that the question information includes a predicted damage improvement amount that is a product of a difference between the predicted damage amount and the predicted damage amount after the countermeasure and the occurrence probability.

Images