Predicted value display method and predicted value display program

Abstract

The objective is to provide a forecast value display method and forecast value display program that improve the readability of forecast values ​​when displaying current and future demand forecasts with confidence probabilities. [Solution] The forecast value display method is a forecast value display method that displays forecast values ​​for demand from the present onward, and is a forecast value display method in which a computer performs the processing of determining the range to display the forecast values ​​according to the thinking tendencies of the user who is viewing the forecast values, and displaying the confidence probability of the forecast values ​​within the determined range.

Description

Technical Field

[0001] This invention relates to a predicted value display method and a predicted value display program.

Background Art

[0002] There are known decision-making tools and decision-making simulation systems for evaluating methods of executing plans for achieving goals (see, for example, Patent Document 1).

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] As a plan for achieving a goal, for example, there is a sales volume plan based on the demand prediction of automobiles by an automobile manufacturer. In demand prediction, if an uncertain event occurs, the demand prediction may deviate. For example, when an event such as a revision of a law that prohibits the installation of lithium-ion batteries in plug-in EVs (Electric Vehicles) occurs in the future, the demand prediction for plug-in EVs equipped with lithium-ion batteries may go awry. As a result, there is a risk that the sales volume target of plug-in EVs cannot be achieved.

[0005] Therefore, automotive manufacturers' representatives are required to make decisions about changing battery suppliers from the perspective of achieving their targets. For example, they may be required to decide to switch from a supplier of lithium-ion batteries to a supplier of nickel-metal hydride batteries. Depending on when the supplier change is made, the target may or may not be achieved. Therefore, the representative conducts simulations to see whether the target can be achieved at several specified change timings, checks the demand forecast displayed in the simulation results, and then decides on the timing of the change.

[0006] However, if, for example, the demand forecast values ​​for each change period are displayed along with their average value and a 95% confidence interval, the readability of the demand forecast values ​​becomes cluttered. In other words, the readability of the demand forecast values ​​decreases. On the other hand, if the average value of the demand forecast values ​​for each change period is displayed alone, the confidence probability of the forecast value is excluded from the display, thus hindering the verification of the reliability of the forecast value.

[0007] Therefore, one objective is to provide a forecast value display method and forecast value display program that improves the readability when displaying forecast values ​​of current and future demand with confidence probabilities. [Means for solving the problem]

[0008] In one embodiment, the forecast value display method is a forecast value display method that displays forecast values ​​for demand from the present onward, and is a forecast value display method in which a computer performs the processing of determining the range to display the forecast values ​​according to the thinking tendencies of the user who is viewing the forecast values, and displaying the confidence probability of the forecast values ​​within the determined range. [Effects of the Invention]

[0009] This improves the readability of forecast demand values ​​displayed with confidence levels. [Brief explanation of the drawing]

[0010] [Figure 1]This is an example of a predicted value display system. [Figure 2] This is an example of the hardware configuration for a prediction value display server. [Figure 3] This is an example of the functional configuration of a prediction value display server. [Figure 4] (a) is an example of a risk tolerance profile. (b) is another example of a risk tolerance profile. [Figure 5] This flowchart shows an example of the operation of the predicted value display server according to the first embodiment. [Figure 6] (a) is an example of a comparative example. (b) is another example of a comparative example. (c) is an example of an example. (d) is another example of an example. [Figure 7] This flowchart shows an example of the operation of the predicted value display server according to the second embodiment. [Figure 8] (a) is a diagram illustrating an example of a risk factor. (b) is a diagram illustrating an example of adjustment. (c) is a diagram illustrating an example of the best time for change for a person with a stable thinking tendency. [Figure 9] This diagram illustrates an example of the relationship between risk factors and risk coefficients. [Figure 10] (a) is a diagram illustrating an example of a risk factor. (b) is a diagram illustrating another example of the relationship between a risk factor and a risk coefficient. (c) is a diagram illustrating an example of the best time to make a change for someone with a speculative mindset. [Modes for carrying out the invention]

[0011] The following will explain the implementation of this project with reference to the drawings.

[0012] (First Embodiment) As shown in Figure 1, the prediction value display system ST is a computer system including a terminal device 10 and a prediction value display server 100. The terminal device 10 and the prediction value display server 100 are connected via a communication network NW. The communication network NW includes either a LAN (Local Area Network) or the Internet, or both.

[0013] In Figure 1, a PC (Personal Computer) is shown as an example of a terminal device 10, but the terminal device 10 is not limited to a PC. The terminal device 10 may also be a smart device such as a smartphone or tablet. In Figure 1, a physical server device is shown as an example of a prediction value display server 100, but the prediction value display server 100 may also be a virtual server device. In Figure 1, one prediction value display server 100 is shown as an example, but multiple prediction value display servers 100 may be provided in the prediction value display system ST, and various processes performed by the prediction value display server 100 may be distributed among the multiple prediction value display servers 100.

[0014] The forecast value display system ST is used by users 11 and 15 belonging to business companies. Users 11 and 15 are examples of users. The businesses to which users 11 and 15 belong may also be users. Business companies may be manufacturers that produce products or non-manufacturers that provide services. Manufacturers include, but are not limited to, automobile manufacturers, electronics manufacturers, etc. Non-manufacturers include, but are not limited to, restaurants, retail stores, etc.

[0015] Both users 11 and 15 can use the prediction value display system ST by operating the input device 12 provided in the terminal device 10 and accessing the prediction value display server 100. For example, when users 11 and 15 individually perform a predetermined operation on the input device 12, the control device 13 of the terminal device 10 transmits an instruction corresponding to the predetermined operation to the prediction value display server 100. When the prediction value display server 100 receives the instruction, it executes various processes required for displaying the prediction value based on the received instruction and transmits the processing result to the control device 13.

[0016] Although details will be described later, for example, when the prediction value display server 100 receives an instruction corresponding to a predetermined operation, it determines a display range for displaying the predicted value of demand after the current time according to the thinking tendency of users 11 and 15. When the prediction value display server 100 determines the display range, it displays a confidence probability corresponding to the confidence interval of the prediction value in the determined display range.

[0017] Referring to FIG. 2, the hardware configuration of the prediction value display server 100 will be described. Since the terminal device 10 described above basically has the same hardware configuration as the hardware configuration of the prediction value display server 100, detailed description thereof will be omitted.

[0018] The prediction value display server 100 includes a CPU (Central Processing Unit) 100A as a processor, a RAM (Random Access Memory) 100B and a ROM (Read Only Memory) 100C as memories. The prediction value display server 100 includes a network I / F (interface) 100D and a HDD (Hard Disk Drive) 100E. Instead of the HDD (Hard Disk Drive) 100E, an SSD (Solid State Drive) may be adopted.

[0019] The predicted value display server 100 may include, as necessary, at least one of the following: input I / F 100F, output I / F 100G, input / output I / F 100H, and drive device 100I. The CPU 100A to the drive device 100I are connected to each other by an internal bus 100J. In other words, the predicted value display server 100 can be implemented by a computer.

[0020] Input I / F 100F is connected to an input device 710. Examples of input devices 710 include keyboards, mice, and touch panels. Output I / F 100G is connected to a display device 720. Examples of display devices 720 include liquid crystal displays. Input / output I / F 100H is connected to a semiconductor memory 730. Examples of semiconductor memory 730 include USB (Universal Serial Bus) memory and flash memory. Input / output I / F 100H reads the predicted value display program stored in the semiconductor memory 730. Input I / F 100F and Input / output I / F 100H are equipped with, for example, USB ports. Output I / F 100G is equipped with, for example, a DisplayPort.

[0021] A portable recording medium 740 is inserted into the drive unit 100I. The portable recording medium 740 can be a removable disk such as a CD (Compact Disc)-ROM or a DVD (Digital Versatile Disc). The drive unit 100I reads the predicted value display program recorded on the portable recording medium 740. The network interface 100D includes, for example, a LAN port and a communication circuit. The communication circuit includes either a wired communication circuit or a wireless communication circuit, or both. The network interface 100D is connected to a communication network NW.

[0022] The CPU 100A temporarily stores the predicted value display program, which is stored in at least one of the ROM 100C, HDD 100E, and semiconductor memory 730, in RAM 100B. The CPU 100A also temporarily stores the predicted value display program, which is recorded in the portable recording medium 740, in RAM 100B. By executing the stored predicted value display program, the CPU 100A realizes various functions described later and also executes a predicted value display method that includes various processes described later. The predicted value display program should conform to the flowchart described later.

[0023] The functional configuration of the predicted value display server 100 will be explained with reference to Figures 3 and 4. Figure 3 shows the main functions of the predicted value display server 100.

[0024] As shown in Figure 3, the predicted value display server 100 comprises a storage unit 110, a processing unit 120, and a communication unit 130. The storage unit 110 can be implemented by either the RAM 100B or the HDD 100E, or both. The processing unit 120 can be implemented by the CPU 100A described above. The communication unit 130 can be implemented by the network I / F 100D described above.

[0025] The memory unit 110, the processing unit 120, and the communication unit 130 are connected to each other. The memory unit 110 includes a setting memory unit 111 and a profile memory unit 112. The processing unit 120 includes a simulation unit 121, a risk coefficient calculation unit 122, a rank assignment unit 123, and a visualization unit 124. The processing unit 120 uses the simulation unit 121, the risk coefficient calculation unit 122, the rank assignment unit 123, and the visualization unit 124 to perform various processes required for displaying predicted values.

[0026] The setting memory unit 111 stores various setting information to be provided to the simulation unit 121. The setting information includes, for example, KPI definition information representing the definition of KPIs (Key Performance Indicators) and graph information that combines knowledge graphs and causal analysis. The setting information also includes action information regarding various actions taken by users 11, 15 or the business company to which users 11, 15 belong, and timing information regarding the timing of any of these actions. The setting information may also include deadline information regarding the timing for determining whether or not the pre-set KPI target values ​​have been achieved.

[0027] For example, if the operating company is an automobile manufacturer, the KPIs would be the annual number of cars sold or the annual sales volume. Also, if the operating company is an automobile manufacturer, the actions taken by the user11,15 or the automobile manufacturer would be to switch from a supplier of lithium-ion batteries to a supplier of nickel-metal hydride batteries. Furthermore, if the operating company is an automobile manufacturer, the events would be such as the amendment of laws prohibiting the installation of lithium-ion batteries in plug-in EVs.

[0028] A knowledge graph, for example, is a graph that visualizes the relationships between data that cause KPIs to fluctuate, thereby supporting information retrieval and inference. Causal analysis, for example, is an analytical method that estimates the causal relationships inherent in the data that cause KPIs to fluctuate. In this way, the setting storage unit 111 stores various setting information to be provided to the simulation unit 121.

[0029] The profile storage unit 112 stores the profile information of users 11 and 15. The profile information represents the thinking tendencies of users 11 and 15 regarding demand forecast risk, where the simulation results produce undesirable predicted values ​​for them. For example, the probability of falling below the target value mentioned above corresponds to demand forecast risk. As shown in Figures 4(a) and (b), the profile information includes either risk tolerance profile R1 or R2. Risk tolerance profiles R1 and R2 are generated in advance based on the self-declarations and past choices of users 11 and 15.

[0030] First, as shown in Figure 4(a), the risk tolerance profile R1 is the profile of user 11 who has a speculative mindset and a preference for high risk, high return. For example, if user 11 has a speculative mindset, the risk tolerance profile R1 shows that as the demand forecast risk increases, user 11's risk tolerance for that demand forecast risk decreases in a gradual curve. In the risk tolerance profile R1, even if the demand forecast risk corresponds to the highest value "1", user 11's risk tolerance remains at a predetermined value "T" which is greater than the lowest value "0". As will be explained in more detail later, demand forecast risk can be identified by risk factors, and risk tolerance can be identified by risk coefficients.

[0031] On the other hand, as shown in Figure 4(b), the risk tolerance profile R2 is the profile of a user with a stable thinking tendency who prefers low risk and low return. For example, if user 15, who is different from user 11, has a stable thinking tendency, the risk tolerance profile R2 shows that as the demand forecast risk increases, user 15's risk tolerance for that demand forecast risk curves sharply downwards. In the risk tolerance profile R2, user 11's risk tolerance reaches the minimum value of "0" before the demand forecast risk reaches the maximum value of "1". The profile storage unit 112 stores profile information for each user that includes either such risk tolerance profiles R1 or R2.

[0032] The simulation unit 121 performs simulations related to the KPI from the present onward. By performing simulations, the simulation unit 121 calculates KPI forecast values ​​that represent the forecast values ​​for demand from the present onward. For example, if the KPI is the annual number of automobiles sold, the simulation unit 121 calculates KPI forecast values ​​that represent the forecast values ​​for automobile demand from the present onward. The simulation unit 121 obtains the setting information required for the simulation from the setting storage unit 111 and performs various simulations based on the obtained setting information. For example, the simulation unit 121 performs simulations for each action period from the present to the decision period represented by the deadline information.

[0033] In addition, the simulation unit 121 acquires profile information and determines the range for displaying the predicted KPI values ​​based on either the risk tolerance profile R1 or R2 included in the profile information. For example, if the profile information includes the risk tolerance profile R1, the simulation unit 121 determines a first range that is above the average of the upper and lower limits of the 95% confidence interval for the predicted KPI values. If the profile information includes the risk tolerance profile R2, the simulation unit 121 determines a second range that is below the average of the upper and lower limits of the 95% confidence interval for the predicted KPI values.

[0034] The risk coefficient calculation unit 122 calculates a risk coefficient that quantifies the demand forecast risk. More specifically, the risk coefficient calculation unit 122 first calculates risk factors for predetermined actions and timings related to the KPI forecast value. The risk coefficient calculation unit 122 can calculate, for example, the variance of the KPI forecast value and the fluctuation (volatility) of the range of the KPI forecast value for predetermined actions and timings, and based on these, it can quantify and calculate the risk factors. Once the risk factors are calculated, the risk coefficient calculation unit 122 normalizes the risk factors for all candidate actions and timings so that the risk factors have a maximum value of "1". As a result, the numerical value of the risk factors is normalized to a range from a minimum value of "0" to a maximum value of "1".

[0035] The risk coefficient calculation unit 122 normalizes the risk factors, obtains profile information, and calculates a risk coefficient for the KPI forecast value based on the relationship between either the risk tolerance profile R1 or R2 included in the profile information and the normalized risk factors. For example, the risk coefficient calculation unit 122 can calculate the risk coefficient by identifying the corresponding value of the risk tolerance profile R1 according to the numerical value of the risk factor. Once the risk coefficient calculation unit 122 calculates the risk coefficient, it adjusts the KPI forecast value by multiplying it by the risk coefficient.

[0036] As a result, the forecast display server 100 can present low-risk, low-return demand forecasts to users 15 with a conservative mindset, without presenting them to users 11 with a conservative mindset. On the other hand, the forecast display server 100 can present high-risk, high-return demand forecasts to users 11 with a conservative mindset, without presenting them to users 15 with a conservative mindset.

[0037] The rank assignment unit 123 assigns ranks to combinations of actions and timing that identify the adjusted KPI forecast values ​​at the time of determining goal achievement. For example, if any of the adjusted KPI forecast values ​​are higher than the remaining adjusted KPI forecast values ​​at the time of determining goal achievement, the rank assignment unit 123 assigns a first-place rank to the combination of actions and timing that identify any of the KPI forecast values. This uniquely determines, for example, the action of changing suppliers and the optimal timing for doing so. In this case, the rank assignment unit 123 may also assign a second-place rank to the combination of actions and timing that identify the highest KPI forecast value among the remaining KPI forecast values.

[0038] The visualization unit 124 displays the confidence probability of the KPI prediction values ​​within the range determined by the simulation unit 121 on the display device 14. For example, the visualization unit 124 displays the upper or lower limit of the 95% confidence interval for the KPI prediction values ​​as the confidence probability on the display device 14. More specifically, when the simulation unit 121 determines the first range described above, the visualization unit 124 displays the upper limit of the 95% confidence interval corresponding to the first range as the confidence probability. When the simulation unit 121 determines the second range described above, the visualization unit 124 displays the lower limit of the 95% confidence interval corresponding to the second range as the confidence probability.

[0039] Referring to Figure 5, the operation of the predicted value display server 100 according to the first embodiment will be described.

[0040] First, the simulation unit 121 acquires setting information (step S1). For example, when user 11 operates the input device 12 and instructs the control device 13 to start running a simulation to forecast demand for KPIs, the simulation unit 121 acquires the setting information.

[0041] Once the configuration information is obtained, the simulation unit 121 obtains profile information (step S2). For example, if user 11 operates the input device 12 and inputs a speculative thinking tendency as user 11's own thinking tendency to the control device 13, the simulation unit 121 obtains profile information including the risk tolerance profile R1 (see Figure 4(a)). Once the simulation unit 121 obtains the profile information, it waits until it detects the input of event information (step S3: NO). The event information is information that includes uncertain events that cause the simulation to change.

[0042] When the simulation unit 121 detects the input of event information (step S3: YES), it executes a simulation (step S4). For example, when user 11 operates the input device 12 and inputs the timing of an event to the control device 13, the simulation unit 121 detects the event information and executes a simulation. For example, the simulation unit 121 executes a simulation to forecast the demand for KPIs based on the KPI definition information, graph information, behavior information, timing information included in the acquired setting information, and the event information.

[0043] Once the simulation is complete, the simulation unit 121 determines the display range (step S5). Specifically, the simulation unit 121 determines the range in which the predicted KPI values ​​will be displayed. The simulation unit 121 can determine the display range based on the acquired profile information. Once the simulation unit 121 has determined the display range, the visualization unit 124 displays the confidence probability of the predicted KPI values ​​within the determined display range (step S6), and then terminates the process.

[0044] Refer to Figure 6 to explain the effects of this case in comparison with the comparative example.

[0045] First, as a comparative example, when the average values ​​of the KPI predictions are displayed without confidence probabilities, the display device 14 will show the average values ​​of the KPI predictions F0m, F1m, F2m, and F3m individually, as shown in Figure 6(a).

[0046] The average KPI forecast value F0m represents the case where an action was taken at the action time "t0" which is closest to the present, during the period from the present to the event occurrence time "t3". The average KPI forecast value F0m branches out and increases from the action time "t0" which is closest to the present, among the KPI forecast values ​​Fx from the present to the event occurrence time "t3". The average KPI forecast value F1m represents the case where an action was taken at the action time "t1" which is the next closest to the present after the action time "t0", during the period from the present to the event occurrence time "t3". The average KPI forecast value F1m branches out and increases from the action time "t1" which is the next closest to the present after the action time "t0", among the KPI forecast values ​​Fx from the present to the event occurrence time "t3".

[0047] The average KPI forecast value F2m represents the case where an action is taken at the next closest action time "t2" after the action time "t1" during the period from the present to the event occurrence time "t3". The average KPI forecast value F2m branches off from the KPI forecast value Fx from the present to the event occurrence time "t3", with the next closest action time "t2" after the action time "t1" as the starting point and rising. The average KPI forecast value F3m represents the case where an action is taken at the event occurrence time "t3". The average KPI forecast value F3m extends upward from the tip of the KPI forecast value Fx.

[0048] Thus, when the average values ​​of the KPI predictions F0m, F1m, F2m, and F3m are each displayed individually on the display device 14, neither users 11 nor 15 can visually perceive the confidence probability of the KPI predictions. In other words, the confidence probability is unknown to users 11 and 15, making it difficult for them to decide when they should take action.

[0049] Next, as another comparative example, when the average values ​​of the KPI predictions are displayed along with their confidence probabilities, as shown in Figure 6(b), the display device 14 displays the average values ​​of the KPI predictions F0m, F1m, F2m, F3m, as well as the confidence probabilities of the KPI predictions F0u, F0l, F1u, F1l, F2u, F2l, F3u, F3l, respectively.

[0050] For example, the confidence probability F0u of the predicted KPI value is a 2.5% probability that it will be above the average predicted KPI value F0m, and corresponds to the upper limit of the 95% confidence interval. The confidence probability F0l of the predicted KPI value is a 2.5% probability that it will be below the average predicted KPI value F0m, and corresponds to the lower limit of the 95% confidence interval. The confidence probabilities F1u, F1l, F2u, F2l, F3u, and F3l of the predicted KPI values ​​are basically the same as the confidence probabilities F0u and F0l, so a detailed explanation is omitted.

[0051] Thus, when the average values ​​F0m, F1m, F2m, F3m of the KPI predictions, as well as the confidence probabilities F0u, F0l, F1u, F1l, F2u, F2l, F3u, F3l of the KPI predictions are displayed on the display device 14, the visibility of the KPI predictions becomes cluttered. In other words, the visibility of the KPI predictions for users 11 and 15 decreases. Furthermore, in this case, at the decision-making period represented by the deadline information, the display range W0 of the average values ​​F0m, ..., F3m of all KPI predictions and the overall confidence probabilities F0u, ..., F3l of all KPI predictions spans a wide range. As a result, both users 11 and 15 may make incorrect decisions about when to take their actions.

[0052] On the other hand, as an example, when the confidence probabilities of the predicted KPI values ​​according to the user 11's thinking tendencies are displayed without the average value, the display device 14 displays the confidence probabilities F0u, F1u, F2u, and F3u of the predicted KPI values, as shown in Figure 6(c). The display range W1 of the confidence probabilities F0u, F1u, F2u, and F3u of the predicted KPI values ​​is narrower and more limited than the display range W0, thus improving the visibility for the user 11. The display range W1 is an example of the first range. Because the user 11 has a speculative thinking tendency, at the decision time represented by the deadline information, it can decide that it is desirable to take action at the action time "t0", which is the factor that calculates the confidence probability F0u of the predicted KPI value that significantly exceeds the target.

[0053] In another embodiment, when the confidence probabilities of the predicted KPI values ​​according to the user 15's thinking tendencies are displayed without the average value, the display device 14 displays the confidence probabilities F0l, F1l, F2l, and F3l of the predicted KPI values, as shown in Figure 6(d). The display range W2 of the confidence probabilities F0l, F1l, F2l, and F3l of the predicted KPI values ​​is narrower and more limited than the display range W0, thus improving the visibility for the user 15. The display range W2 is an example of a second range. Because the user 15 has a stable thinking tendency, they can decide that it is desirable to take action at the action timing "t2", which is the factor for calculating the confidence probability F2l of the predicted KPI value that is closest to the goal, at the decision timing represented by the deadline information.

[0054] Thus, according to the first embodiment, the forecast value display server 100 determines the display range for displaying KPI forecast values ​​according to the thinking tendencies of users 11 and 15 who are viewing the KPI forecast values, which are demand forecasts for the present and beyond, and displays the confidence probability of the KPI forecast values ​​within the determined display range. This improves the readability when displaying KPI forecast values ​​with confidence probabilities.

[0055] (Second Embodiment) The second embodiment of this invention will be described with reference to Figures 7 to 10. In Figure 7, the same reference numerals are used for processes similar to those described with reference to Figure 5, and their detailed descriptions are omitted. In the second embodiment, by adjusting the average value of the above-mentioned KPI prediction values, a time is provided to identify the best KPI prediction values ​​according to the thinking tendencies of users 11 and 15.

[0056] First, as shown in Figure 7, in the process of step S5, once the simulation unit 121 determines the display range, the risk coefficient calculation unit 122 calculates the risk factors (step S11). As described in the first embodiment, the risk coefficient calculation unit 122 calculates the variance of the KPI prediction values ​​and the fluctuation (volatility) of the range of the KPI prediction values ​​for predetermined actions and timings, and based on these, it can calculate quantified risk factors.

[0057] Once the risk factors are calculated, the risk coefficient calculation unit 122 normalizes the risk factors (step S12) and calculates the risk coefficient (step S13). More specifically, the risk coefficient calculation unit 122 calculates the risk coefficient for the KPI forecast value based on the relationship between either the risk tolerance profile R1 or R2 included in the profile information and the normalized risk factors.

[0058] For example, as shown in Figure 8(a), if a risk factor of "0.3" is calculated for the average KPI forecast value F0m, then, as shown in Figure 9, if the risk tolerance profile R2 is used, the risk coefficient calculation unit 122 can calculate a risk coefficient of "0.5". In the risk tolerance profile R2, the risk factor of "0.3" corresponds to a risk coefficient of "0.5". Also, as shown in Figure 8(a), if a risk factor of "0.1" is calculated for the average KPI forecast value F2m, then, as shown in Figure 9, if the risk tolerance profile R2 is used, the risk coefficient calculation unit 122 can calculate a risk coefficient of "0.95". In the risk tolerance profile R2, the risk factor of "0.1" corresponds to a risk coefficient of "0.95". In this way, the risk coefficient calculation unit 122 can calculate the risk coefficient by identifying the corresponding value in the risk tolerance profile R2 according to the numerical value of the risk factor.

[0059] As shown in Figure 7, once the risk coefficient is calculated, the risk coefficient calculation unit 122 adjusts the average value of the KPI prediction values ​​(step S14). For example, if the risk coefficient calculation unit 122 calculates a risk coefficient of "0.5", as shown in Figure 8(b), the average value of the KPI prediction values ​​F0m before applying the risk coefficient of "0.5" is multiplied by the risk coefficient of "0.5" to adjust the average value of the KPI prediction values ​​F0m. As a result, the average value of the KPI prediction values ​​F0m before applying the risk coefficient of "0.5" is adjusted to the average value of the KPI prediction values ​​F0M after applying the risk coefficient of "0.5". Although not shown, the risk coefficient calculation unit 122 similarly adjusts the average value of the KPI prediction values ​​F2m based on a risk coefficient of "0.95".

[0060] As shown in Figure 7, when the risk coefficient calculation unit 122 adjusts the average value of the KPI forecast, the rank assignment unit 123 assigns a rank (step S15). The risk coefficient calculation unit 122 can assign a rank to the combination of actions and timing that identify the adjusted average value of the KPI forecast. For example, as shown in Figure 8(c), when the average values ​​of the KPI forecast F0m and F2m are adjusted to the average values ​​of the KPI forecast F0M and F2M, respectively, the rank assignment unit 123 assigns a rank to the combination of actions and timing that identify the average values ​​of the KPI forecast F0M and F2M.

[0061] In this case, the ranking unit 123 assigns a first rank to the combination of action and action timing "t2", which is a factor in calculating the average KPI forecast value F2M that exceeds the target. The ranking unit 123 also assigns a second rank to the combination of action and action timing "t0", which is a factor in calculating the average KPI forecast value F0M that falls below the target. As a result, for example, the best time to change suppliers is identified as action timing "t2" for a user 15 with a stable thinking tendency. Once the ranking unit 123 assigns ranks, the visualization unit 124 executes the process in step S6, as shown in Figure 7, and terminates the process.

[0062] As shown in Figure 10(a), if a risk factor of "0.3" is calculated for the average KPI forecast value F0m, then, as shown in Figure 10(b), if the risk tolerance profile is R1, the risk coefficient calculation unit 122 can calculate a risk coefficient of "0.98". In the risk tolerance profile R1, the risk factor of "0.3" corresponds to a risk coefficient of "0.98". Also, as shown in Figure 10(a), if a risk factor of "0.1" is calculated for the average KPI forecast value F2m, then, as shown in Figure 10(b), if the risk tolerance profile is R1, the risk coefficient calculation unit 122 can calculate a risk coefficient of "0.99". In the risk tolerance profile R1, the risk factor of "0.1" corresponds to a risk coefficient of "0.99".

[0063] As a result, as shown in Figure 10(c), in the case of risk tolerance profile R1, unlike risk tolerance profile R2, the average values ​​of the predicted KPIs F0m and F2m are adjusted to the average values ​​of the predicted KPIs F0p and F2p, respectively.

[0064] In this case, the rank assignment unit 123 assigns a first rank to the combination of action and action timing t0, which is a factor in calculating the average KPI forecast value F0p that significantly exceeds the target. The rank assignment unit 123 also assigns a second rank to the combination of action and action timing t2, which is a factor in calculating the average KPI forecast value F2p that slightly exceeds the target. As a result, for example, the best time to change suppliers is identified as action timing "t0" for user 11, who has a speculative thinking tendency. Thus, according to the second embodiment, the best time to change is provided according to the thinking tendencies of each user 11 and 15.

[0065] In the second embodiment, the average value of the predicted KPI was described as being adjusted. However, by adjusting the confidence probability of the predicted KPI, a timing for identifying the best predicted KPI value according to the thinking tendencies of users 11 and 15 may be provided. That is, by adjusting the confidence probabilities F0u, F1u, F2u, F3u of the predicted KPI (see Figure 6(c)), a timing for identifying the best predicted KPI value according to the thinking tendencies of user 11 may be provided. Also, by adjusting the confidence probabilities F0l, F1l, F2l, F3l of the predicted KPI (see Figure 6(d)), a timing for identifying the best predicted KPI value according to the thinking tendencies of user 15 may be provided.

[0066] Although preferred embodiments of the present invention have been described in detail above, the invention is not limited to specific embodiments, and various modifications and changes are possible within the scope of the gist of the invention as described in the claims. For example, in the embodiments described above, the upper and lower limits of the 95% confidence interval were explained as an example of confidence probability, but instead of the 95% confidence interval, the upper and lower limits of the 90% confidence interval or the upper and lower limits of the 99% confidence interval may be applied to the confidence probability in this case.

[0067] Furthermore, the following additional information is disclosed regarding the above explanation. (Note 1) A method for displaying forecast values ​​for current and future demand, wherein a computer performs a process to determine the range in which the forecast values ​​are to be displayed according to the thinking tendencies of the user viewing the forecast values, and displays the confidence probability of the forecast values ​​within the determined range. (Note 2) The method for displaying predicted values ​​according to Note 1, characterized in that, if the user's thinking tendency is a speculative first thinking tendency, a first range is determined as the range that exceeds the average value of the predicted values, and the confidence probability of the first predicted values ​​in the determined first range is displayed. (Note 3) The method for displaying forecast values ​​according to Note 1, characterized in that, if the user's thinking tendency is a speculative first thinking tendency with respect to the demand forecast risk that will result in forecast values ​​undesirable to the user, a first range is determined that exceeds the average value of the forecast values, the first forecast values ​​of the determined first range are adjusted based on information representing the relationship between the level of the demand forecast risk and the user's tolerance for the demand forecast risk, and the confidence probability of the adjusted first forecast values ​​of the first range is displayed. (Note 4) The method for displaying predicted values ​​according to Note 1, characterized in that, if the user's thinking tendency is a stable second thinking tendency, a second range is determined as the range that is below the average value of the predicted values, and the confidence probability of the second predicted values ​​in the determined second range is displayed. (Note 5) The method for displaying forecast values ​​according to Note 1, characterized in that, if the user's thinking tendency is a stable second thinking tendency with respect to the demand forecast risk that results in undesirable forecast values ​​for the user, a second range is determined as the range that is below the average value of the forecast values, the second forecast values ​​of the determined second range are adjusted based on information representing the relationship between the level of the demand forecast risk and the user's tolerance for the demand forecast risk, and the confidence probability of the adjusted second forecast values ​​of the second range is displayed. (Note 6) The method for displaying predicted values ​​according to any one of Notes 1 to 5, characterized in that the confidence probability corresponds to one of a 95% confidence interval, a 90% confidence interval, or a 99% confidence interval. (Note 7) A forecast value display program that displays forecast values ​​of demand from the present onward on a computer, wherein the program causes the computer to perform the following processes: determine the range in which to display the forecast values ​​according to the thinking tendencies of the user viewing the forecast values, and display the confidence probability of the forecast values ​​within the determined range. [Explanation of Symbols]

[0068] ST Prediction Value Display System 100 Prediction Value Display Server 121 Simulation Department 122 Risk coefficient calculation unit 123 Rank Assignment Section 124 Visualization Section R1, R2 Risk Tolerance Profile

Claims

1. A method for displaying forecast values ​​that shows forecast values ​​for demand from the present onward, The range in which the predicted values ​​are displayed is determined according to the thinking tendencies of the user who is viewing the predicted values. The confidence probability of the predicted value within the determined range is displayed. A method for displaying predicted values ​​performed by a computer.

2. If the user's thinking tendency is a speculative first thinking tendency, a first range is determined that exceeds the average value of the predicted values. The confidence probability of the first predicted value within the determined first range is displayed. The method for displaying predicted values ​​according to claim 1.

3. If the user's thinking tendency is a speculative first thinking tendency with respect to the demand forecast risk that will result in undesirable forecast values, then a first range is determined that exceeds the average value of the forecast values. The first forecast value within the determined first range is adjusted based on information representing the relationship between the level of the demand forecast risk and the user's tolerance for the demand forecast risk. The confidence probability of the first predicted value within the adjusted first range is displayed. The method for displaying predicted values ​​according to claim 1.

4. If the user's thinking tendency is a stable second thinking tendency, a second range is determined that is below the average value of the predicted values. The confidence probability of the second predicted value within the determined second range is displayed. The method for displaying predicted values ​​according to claim 1.

5. If the user's thinking tendency is a stable second thinking tendency with respect to the demand forecast risk of producing undesirable forecast values, then a second range below the average value of the forecast values ​​is determined as the range. The second forecast value within the determined second range is adjusted based on information representing the relationship between the level of the demand forecast risk and the user's tolerance for the demand forecast risk. The confidence probability of the second predicted value within the adjusted second range is displayed. The method for displaying predicted values ​​according to claim 1.

6. A forecast value display program that displays forecast values ​​for demand from the present onward on a computer, The range in which the predicted values ​​are displayed is determined according to the thinking tendencies of the user who is viewing the predicted values. The confidence probability of the predicted value within the determined range is displayed. A program that displays predicted values ​​and causes the computer to perform the processing.

Images