JP5789718B2 - Risk scenario analysis system and risk scenario analysis method - Google Patents

Description

  The present invention relates to a risk evaluation method for business cash flow.

  As a background art of this technical field, there is known a method for predicting a business cash flow based on a combination of external fluctuation factors related to business profit. For example, Japanese Patent No. 4137537 discloses a method of calculating and displaying a cash flow as a probability distribution by calculating a current value and a fluctuation rate of external fluctuation factors as a premise. A general method is to calculate the minimum and maximum cash flow values from the combination of the minimum and maximum values of external fluctuation factors.

Japanese Patent No. 4137537

  When there are N types of external fluctuation factors when using the current value and the fluctuation rate, or the minimum value and the maximum value, these combinations will be 2 to the Nth power, but what value can occur with what probability? I can't know. Thus, it is considered that the probability distribution of cash flow fluctuation is obtained while grasping which value can occur with what probability by treating the external fluctuation factor with the probability distribution. However, when there are N types of external variation factors and the interval resolution of the probability distribution is M stages, the combination of external variation factors, that is, the number of scenarios, is an enormous number of orders of N to the Mth power. For example, when there are 10 types of variation factors and the section resolution is 10 stages, the number of scenarios that were conventionally 1024 is 10 billion. Therefore, there is a problem that it is practically impossible to identify a specific scenario from the cash flow, which is necessary for the management to present materials for making a management decision.

  As one means for solving the above problem, a risk scenario analysis system for analyzing a relationship between external flow factor information related to cash flow and a cash flow value, a database storing probability distribution information of external variable factor information, Based on the probability distribution information, a cash flow probability distribution calculation unit that calculates cash flow value probability distribution information, and from the calculated cash flow probability distribution information, calculates external variation factor combination information corresponding to the cash flow value. There is a risk scenario analysis system including an external variation scenario calculation unit and an output unit that outputs the calculated combination information on a screen.

  By using the above method, external factors that affect the cash flow of the business are handled with a probability distribution, so that it is possible to obtain a probability distribution of cash flow fluctuations while grasping which event occurs with what probability. Further, by selecting a cash flow value and adjusting the huge number of scenarios such as what combination of external factors the value is caused by clustering, the user can easily discriminate.

The figure which showed the flow of the basic process example of this invention. The figure which showed the example of probability distribution of an external fluctuation factor. The figure which showed the example of the external fluctuation factor setting part and cash flow probability distribution display. The figure which showed the example of a screen display of an external fluctuation factor related setting part. The figure which showed the example of the correspondence scenario display. The figure which showed the example of a scenario display of a map format. The figure which showed the example of a cluster type scenario display. The figure which showed the example of a scenario display of the table format with respect to the value of the cash flow designated individually. The figure which showed the example of a scenario display of the map format with respect to the value of the cash flow designated individually. The figure which showed the example of a cluster type scenario display with respect to the value of the cash flow designated individually. The figure which showed the example of a display of a bar graph format. The figure which showed the example of the setting of various variables. The figure which showed the example which applied this invention to the water infrastructure business. The figure which showed the structural example of this invention.

  In this embodiment, an example of the risk scenario display method 100 will be described. FIG. 14 shows a configuration diagram of the present invention. A database unit 901 stores a probability distribution of external variation factors related to the business. Reference numeral 902 denotes a program of the present invention stored in a storage medium or the like, which is transferred to the memory unit 903 and executed when it is activated. In addition, the setting by the user is received by the input unit 904 such as a mouse or a keyboard, and the calculation result is displayed on the display or the like by the display unit 905.

  FIG. 1 shows a flowchart of basic processing of this embodiment. After starting the program 902, in 110, the data 111 such as the time series of external fluctuation factors stored in the memory 901 is read, the frequency distribution of these values is calculated, and the probability distribution of the external fluctuation factors related to the cash flow of the business Is generated. Alternatively, the input unit 904 accepts user input, and the external variation factor setting unit 112 corrects the probability distribution of the external variation factor. At 120, the probability distribution of cash flow is calculated from the probability distribution of these external variation factors and stored in the database 901. The input from the user is accepted through the input unit 904, and the external variation factor relation setting unit 121 sets the relationship that the external variation factor gives to the cash flow. Next, at 130, the cash flow probability distribution is displayed on the screen. In 140, the scenario corresponding to the cash flow value is searched. The search method is set by receiving a user input in the search scenario setting unit 141. Finally, the corresponding scenario searched in 150 is displayed on the screen.

  FIG. 2 shows an example of the probability distribution of external variation factors. The probability distribution {Pa} of the external variation factor A is a set of probabilities of occurrence for each interval of values that the variation factor A can take, and the total value is 100%. In the example of FIG. 2, A takes a value in the range of 0.1 to 1.4, for example, the probability Pa (0.1, 0.2) taking a value of 0.1 to 0.2 is 1%. As a result of calculating other external fluctuation factors and a plurality of external fluctuation factors, cash flow fluctuations are also handled in the same format.

  The external variation factor setting unit 112 and the cash flow probability distribution display 130 will be described with reference to FIG. The left side is an external factor variation probability distribution setting area, and in 201, the mode is switched between a probability distribution setting mode and a scenario display mode to be described later. Reference numeral 202 denotes a bar graph representing the probability distribution of external variation factors and the probability distribution obtained by these calculations. For example, a bar graph of an external variation factor is designated with a mouse or the like, and the probability of the designated section is set by dragging up and down while holding down the button. The increase / decrease at this time is distributed according to the probability of other sections and adjusted so that the total is 100%. At this time, the specified external variation factor and the probability distribution other than the distribution used for the calculation are darkly displayed so that the relationship is highlighted. In the example of FIG. 3, B is designated, and the distributions other than the related distributions A + B and Cx (A + B) are darkly displayed. Further, an external variation factor is added by pressing an add button 203, and the selected external variation factor is deleted by pressing a delete button 204 after selecting the external variation factor. Further, past data 111 such as time series is read by a data reading button 205, and a probability distribution is calculated and displayed. A range and a section of the probability distribution are set by a range setting button 206. Reference numeral 220 denotes a probability distribution of cash flow fluctuations calculated from external fluctuation factors. The calculation method will be described in detail below.

  FIG. 4 is a screen display example of the external variation factor related setting unit 121. Reference numerals 231 to 236 denote external variation factors, and these relationships are designated with a mouse or the like. For example, 240 is displayed when external variation factors A and B are associated. First, when the mouse button is pressed when the cursor is over the box 231 and when the mouse button is released when the cursor is over the box 232, the external variation factors A and B are associated to display 240 boxes. To do. Further, by selecting 240, an external variation factor calculation method such as addition, subtraction, multiplication, division, or a specific function is designated. Reference numeral 241 represents a further addition of an external variation factor C. These operations are repeated to obtain the final cash flow 250, and 260 is pressed to complete the setting.

Here, the calculation method of the probability distribution of the external variation factor will be described. Assume that the probability distribution {Q} of f (A, B) is obtained from the probability distributions {Pa} and {Pb} of the external variation factors A and B. The probability of {Pa} for N intervals is Pa (a (0), a (1)), Pa (a (1), a (2)), .. Pa (a (N-1), a ( N)), Pb (b (0), b (1)), Pb (b (1), b (2)), .. Pb (b (M-1 ), b (M)). Q (q (0), q (1)), Q (q (1), q (2)), .. Q (q (T-1), q (T )) Is obtained as follows.
(Equation 1)
Q (q (t), q (t + 1)) = ΣmΣn (Pa (a (m), a (m + 1)) * Pb (b (n), b (n + 1)),
However, q (t) ≦ f (a (m), b (n)) <q (t + 1)). For example, when the calculation method is addition, the above f (a (m), b (n)) is calculated as a (m) + b (n).

  Next, the corresponding scenario display 150 will be described with reference to FIG. The left part of FIG. 5 is a scenario display area when the scenario display is selected in 201, and the right side is a display area of the probability distribution of cash flow fluctuation. Reference numeral 301 denotes a selection part of a scenario display format, which is selected from a table format, a map format, and a cluster format. Other formats may be used. Reference numeral 302 denotes an example in the case where a table format is selected, in which combinations of values of external fluctuation factors corresponding to the bar graph of the probability distribution of cash flow 303 and the occurrence probabilities at that time are displayed.

  FIG. 6 shows an example of scenario display in a map format. The value of the external variation factor corresponding to each value of the cash flow 303 is displayed in color or shape as indicated by 305. Reference numeral 306 denotes a legend. The external variation factors corresponding to the vertical axis and the horizontal axis can be changed by selecting 307 and 308.

  FIG. 7 shows an example of a cluster type scenario display. In addition to the map format shown in FIG. 6, a group of the same cash flow values is displayed. As shown at 310, the close combinations of external variation factors are surrounded by a line, and each cluster is presented. This presentation method makes it possible to globally understand what combination of external factors each cash flow results from. For example, a scenario that takes the third value of cash flow is displayed as distinguished from cluster 4 and cluster 5. For clustering, when the number of external fluctuation factors is N, the values of the external fluctuation factors are normalized by a method such as linear correction such that the maximum value and the minimum value are 1 and 0, respectively. The combination is defined as an N-dimensional vector on the N-dimensional space. Next, when the distance of the vector which takes the value of the same cash flow is below a fixed value L, it defines as the same cluster. Here, the larger the L, the smaller the number of clusters. It is also possible to automatically estimate L by specifying the number of clusters. It also affects the number of intervals and the number of clusters in the probability distribution of cash flow. These values can be specified by the user as will be described later.

  So far, we have described how to display scenarios for all cash flows. However, it is also important to analyze the scenario in detail, for example, by identifying cash flow values that could seriously affect business operations. Hereafter, the display method when individually designated will be described.

  FIG. 8 shows a scenario display example in a tabular format for individually designated cash flow values. When the individually designated check box 401 is designated from FIG. 5, the individual setting mode is set. In the individual designation mode, a bar graph is added to the scenario display format choices as indicated at 410. Here, when a specific value is selected from the graph 402 of the probability distribution of cash flow fluctuation, a combination of related external fluctuation factors is displayed as indicated by 403.

  9 and 10 show display examples of the map format and cluster format in the individual designation mode. FIG. 9 shows a scenario display example in a map format for individually designated cash flow values. When a specific value is selected from the cash flow fluctuation probability distribution graph 402, a combination of related external fluctuation factors is displayed as indicated by 403. FIG. 10 shows an example of a cluster type scenario display for individually designated cash flow values. When a specific value is selected from the cash flow fluctuation probability distribution graph 402, a combination of related external fluctuation factors and a clustered scenario are displayed as indicated by 403.

  FIG. 11 shows a display example in a bar graph format. When a certain value is selected from the cash flow fluctuation probability distribution 402, the value of each external fluctuation factor resulting from this value is highlighted in the bar graph of the probability distribution. With this display format, when a certain cash flow value is taken, it is possible to overlook the degree of occurrence probability of each external variation factor.

  The setting of various variables will be described with reference to FIG. When the setting button 501 arranged in the display area is pressed, a screen as shown in FIG. 12 is displayed. Reference numeral 510 denotes a setting box for the number of sections of cash flow described in the description of FIG. Reference numerals 511 and 512 denote selection buttons for selecting the number of clustering as a distance threshold value or directly as a number. Reference numerals 515, 516, and 517 are options for the probability distribution display method. This is applied when the probability distribution of cash flow fluctuations and external fluctuation factors are represented by probability distributions as shown in FIGS. 515 displays the probability distribution described so far. When 516 is selected, the probability distribution Pr that occurs even once in the period R is displayed. If the probability distribution of each interval of external fluctuation factors is P (a (i), a (i + 1)), Pr (a (i), a (i + 1)) = 1-(1-P (a (i), a (i + 1))) ^ R The period R is set at 518. Further, when 517 is selected, a probability distribution Prs that occurs S times or more in the period R is displayed. Prs is obtained as shown in Equation 2 below, and R and S are set at 519 and 520. When the end button 530 is pressed, the setting of various variables is ended.

  FIG. 13 shows a specific example in which the above method is applied to a water infrastructure project. The power consumption and chemical usage of each device in the water purification plant and seawater desalination plant are determined from the amount of raw water taken and the quality of the raw water. I want. From these, electricity charges, and chemical prices, power costs and chemical costs are calculated, and the running cost of the entire plant is obtained. In addition, personnel expenses are calculated from changes in the number of employees and average wages, and interest payments are determined from changes in borrowings and interest rates. On the other hand, the amount of water sold is determined from the amount of water produced and the yield, and this is multiplied by the water sales price to obtain revenue. Furthermore, the total income is obtained by adding subsidies and other income. By subtracting the total expenditure from the total revenue, the probability distribution of the final cash flow fluctuation can be obtained.

100 Risk scenario display method 110 External variation factor probability distribution 111 Data 112 External variation factor setting 120 Cash flow probability distribution calculation 121 External variation factor related setting 140 Corresponding scenario search 150 Corresponding scenario display 203 Add button 204 Delete button 205 Data load button 206 Range Setting button 301 Scenario display format 401 Check box 410 Scenario display format 501 Setting button 510 Number of sections setting box 901 Database unit 902 Program 903 Memory unit 904 Input unit 905 Display unit

Claims (15)

A risk scenario analysis system that analyzes the relationship between cash flow values and external variable information about cash flows,
A database storing probability distribution information of the external variation factor information;
A cash flow probability distribution calculating unit for calculating probability distribution information of cash flow values based on the probability distribution information;
An external variation scenario calculation unit that calculates combination information of external variation factors corresponding to cash flow values from the calculated cash flow probability distribution information, and an output unit that outputs the calculated combination information on a screen. Risk scenario analysis system characterized by including. The risk scenario analysis system according to claim 1,
It further includes an external variation factor probability distribution calculation unit that receives input of time series data of external factor information, calculates probability distribution information of external factor information from the received time series data, and stores the probability distribution information in the database. Risk scenario analysis system. The risk scenario analysis system according to claim 1,
The risk scenario analysis system further comprising an external variation factor relation setting unit for setting a relationship between the external variation factor information and the cash flow value. The risk scenario analysis system according to claim 3,
The external variation factor relation setting unit receives an input of a cash flow value calculation rule using the external variation factor information, and sets the relevance between the external variation factor information and the cash flow value. Risk scenario analysis system. The risk scenario analysis system according to claim 1,
The output unit receives an input of output format information for selecting an output format of the combination information, and displays the combination information on the screen based on the output format information that has received the input. Analysis system. The risk scenario analysis system according to claim 5,
The risk scenario analysis system, wherein the output format information includes any or all of a table format, a map format, and a cluster format. The risk scenario analysis system according to claim 5,
When the output format information is a cluster format, the external variation scenario calculation unit vectorizes the combination information as a vector of dimensions of the number of external variation factor information, and the distance between the vectors taking the same cash flow value is determined in advance. A risk scenario analysis system characterized in that the combination information is clustered as the same cluster when it is below a predetermined threshold. The risk scenario analysis system according to claim 7,
The risk scenario analysis system, wherein the external variation scenario calculation unit receives an input of the threshold value. The risk scenario analysis system according to claim 7,
The external variation scenario calculation unit receives an input of a maximum value of the number of clusters, and performs clustering so that the number of clusters is the same as the maximum value of the input. The risk scenario analysis system according to claim 2,
The risk scenario analysis system, wherein the external variation factor probability distribution calculation unit receives an input of a period corresponding to the calculated probability distribution and the number of events, and calculates the probability distribution. A risk scenario analysis method for analyzing a relationship between external flow factor information relating to cash flow and a cash flow value using a database storing probability distribution information of the external variable factor information,
A cash flow probability distribution calculating step of calculating a probability distribution information of a cash flow value by a cash flow probability distribution calculating unit based on the probability distribution information;
An external variation scenario calculation step of calculating external variation factor combination information corresponding to the cash flow value by an external variation scenario calculation unit from the calculated cash flow probability distribution information, and outputting the calculated combination information to the screen An output process to be output by the unit,
A risk scenario analysis method characterized by including: A risk scenario analysis method according to claim 11, comprising:
External variation factor probability distribution calculation step of accepting input of time series data of external factor information, calculating probability distribution information of external factor information from the received time series data, and storing it in the database by an external variation factor probability distribution calculation unit A risk scenario analysis method characterized by further comprising: A risk scenario analysis method according to claim 11,
The risk scenario analysis method further comprising an external variable factor relation setting step of setting an association between the external variable factor information and the cash flow value by an external variable factor relation setting unit. A risk scenario analysis method according to claim 13, comprising:
In the external variation factor relation setting step, the external variation factor relationship setting unit receives an input of a cash flow value calculation rule using the external variation factor information, and the external variation factor information and the cash flow value A risk scenario analysis method characterized by setting relevance. A risk scenario analysis method according to claim 11, comprising:
In the output step, the output unit receives an input of output format information for selecting an output format of the combination information, and displays the combination information on the screen based on the output format information that has received the input. A characteristic risk scenario analysis method.

Images