JP2015135617A - Method for evaluation of total performance of forge-proof printed matters and system for evaluation of total performance of forge-proof printed matters - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a system for evaluation of total performance of forge-proof printed matters that, in performing a total evaluation of the forge-proof printed matters with setting multiple requirements to be required for the forge-proof printed matters as evaluation items, has accuracy, rationality and reliability to a quantitative evaluation result of each evaluation item, and to provide a method therefor.SOLUTION: A system has a control unit as a core, and roughly consists of an evaluation tree structure setting unit, a data input unit and an evaluation output unit. The control unit includes an evaluation processing section that consists of a weight coefficient calculation part, a consistency index calculation and determination part, a performance weight coefficient set calculation part, a partial performance evaluation element value calculation part, a partial performance evaluation value calculation part and a total performance evaluation value calculation part. The evaluation processing section evaluates forge-proof printed matters using: evaluation items of a tree structure set by the evaluation tree structure setting unit; and data input by the data input unit.

Description

The present invention relates to a method and system for comprehensively evaluating various requirements such as safety, convenience, social acceptability, and cost required for forgery prevention printed matter.

The role of anti-counterfeit printed matter such as banknotes and passports is extremely important in financial transactions and personal authentication situations. Counterfeit banknotes will lead to the loss of credit in the national economy, and counterfeit passports will serve as an opportunity for heavy criminals such as international terrorists and human traffickers. For this reason, it is required to fully evaluate the safety and ease of use of these forgery-preventing printed materials and firmly ensure the requirements.

The requirements for anti-counterfeited printed materials are very different from those for general commercial printed materials such as books, flyers, and wrapping paper. Ease of use, ease of handling, universal design, durability (physical and chemical), suitability for mechanical processing, manufacturing costs, etc.

Among these evaluation items, there are many quantitative evaluation criteria for the evaluation of physical and chemical durability requirements, and a formalized specific evaluation is practiced. On the other hand, it is difficult to objectively and quantitatively evaluate a method for evaluating the satisfaction degree of safety requirements such as anti-counterfeiting, social acceptability requirements such as aesthetics and universal design. Therefore, the evaluation of these requirements is mainly performed based on the tacit knowledge of the individual, that is, subjective evaluation. From such a background, it is required to objectively evaluate the overall performance of requirements required for anti-counterfeit printed matter. In addition, as a basis for determining the final specifications of a product, the satisfaction of all requirements such as safety, durability, and cost is considered from an overall perspective, and the overall performance of the product specifications is evaluated. It is requested to do.

Although it is different from the evaluation of the total performance of anti-counterfeit printed matter, as a method of comprehensively evaluating certain requirements, the office building owner reviews the contents of the renewal and the end user selects a housing complex that meets his needs A facility performance evaluation system for supporting such decision making has been proposed (for example, see Patent Document 1). In the technique of Patent Document 1, a hierarchical analysis method is used. By using this method, it is considered that a rough comprehensive evaluation of a forgery-preventing printed material can be performed to a certain degree. Hierarchical analysis method is also called Hierarchical Decision Making Method or AHP (Analytic Hierarchy Process), and it is organized into a hierarchical structure of objectives or goals, evaluation criteria for realizing them, and alternatives to be compared in evaluation. , Arithmetically derive the relative importance by pair-wise comparison (take out two elements from multiple elements and perform one-to-one comparison) between elements of each hierarchy, and substitute from the magnitude of the importance This is a technique for comprehensively evaluating or selecting a plan (for example, Non-Patent Document 1). The technique of Patent Document 1 is capable of roughly and simply evaluating the performance of a facility without difficulty and cost while using this hierarchical analysis method and displaying the results in an easy-to-understand manner. Is a feature and can be used even by users who do not have specialized knowledge. Another characteristic is that different values in facility performance can be reflected in the evaluation.

Further, an apparatus for evaluating an alternative to a target using a hierarchical analysis method as in the technique of Patent Document 1 has been proposed (see, for example, Patent Document 2). In the hierarchical analysis method, when an alternative is added or deleted, it is necessary to evaluate again by performing a pair comparison. However, in the technique of Patent Document 2, when an alternative is added or deleted, the pair is not compared again. For the purpose of evaluation, the alternative value is evaluated by converting the quantitative value for the evaluation item by a certain conversion formula and performing the processing of the weighting coefficient and the multiplication. As a result, even when the alternatives are changed, the alternatives can be evaluated without making another pair comparison.

Here, an example of a comprehensive evaluation method for forgery-preventing printed matter using the hierarchical analysis method used in Patent Literature 1 and Patent Literature 2 will be described below. First, as shown in FIG. 1A, an evaluation tree structure in which “counterfeit prevention printed matter with excellent overall performance” is arranged at the top is set, and then the forgery prevention printed matter with excellent overall performance is set. As evaluation items to be realized, “safety”, “convenience”, “social acceptability”, and “cost” are set in the lower hierarchy (see, for example, Non-Patent Document 2). The evaluation item in the hierarchical structure shown in FIG. 1A is an example set based on artifacts metrics: security research and development trends and issues in the anti-counterfeiting technology of Non-Patent Document 2. Accordingly, the hierarchical structure may be set by adding or deleting the evaluation items of the anti-counterfeit printed matter. Here, the hierarchical structure shown in FIG. 1 (a) will be described, but the hierarchical structure of the hierarchical analysis method is not limited to the two-layer structure shown in FIG. 1 (a), but as shown in FIG. 1 (b). In addition, a three-layer structure may be set, or a lower hierarchy may be provided.

Next, a pairwise comparison is performed to determine the importance of each evaluation item when performing comprehensive evaluation, that is, a weighting factor. In order to implement this, the degree of importance shown in Table 1 (pair comparison value) is used.

(table 1)

  Table 2 shows the results of a paired comparison in terms of safety, convenience, social acceptability, and cost, which are lower evaluation items, using the paired comparison values shown in Table 1. The paired comparison values shown in Table 2 show that safety is “somewhat important (3)” compared to convenience, safety is “important (5)” compared to social acceptability, and safety is cost. As a result of comparison, a pairwise comparison value is input in the horizontal column (in the row direction), such as “pretty important (7)”. On the other hand, 1/3, 1/5, and 1/7, which are reciprocals of each paired comparison value, are input in the vertical column (column direction) of safety. Equally important (1) ”is input. Here, an example using a paired comparison value shown in Table 1 will be described. For example, in the case of an evaluation between “very important (9)” and “very important (7)”, the intermediate value is “8”. Can also be used. In addition, when multiple evaluators perform paired comparisons, the variation in evaluation can be reduced by using the geometric mean value of the paired comparison values of each individual in order to reduce the variation in the paired comparison values by the evaluators. Can do. Note that the number of evaluation items to be compared by paired comparison is preferably within seven. This is because if the number exceeds this value, it is difficult to ensure consistency when evaluating each evaluation item with an overall scale.

(Table 2)

Next, a weighting factor is calculated based on the input paired comparison table. Looking at the paired comparison table shown in Table 2 from the top, the safety row is the first row and the convenience is the second row. Similarly, from the left, the safety column is the first row and the convenience row is the second row. As an eye, the paired comparison table shown in Table 2 is a square matrix with 4 rows and 4 columns. Thus, the pair comparison table in the m evaluation items (m is a natural number of 2 or more, and m is 4 in the case of Table 2) can be expressed as a square matrix A having m rows and m columns. Here, the importance of the evaluation item i compared with the evaluation item j is represented as a paired comparison value aij. Similarly, the importance of the evaluation item j compared with the evaluation item i is represented as a paired comparison value aji. At this time, aij = 1 / aji. As a result, the m-by-m paired comparison matrix A in the m evaluation items is (Equation 1). The values of a11... Aii... Amm, which are diagonal elements in the paired comparison matrix A, are all 1.

・・・（１） (Equation 1)

... (1)

Next, based on the paired comparison matrix A, the weight coefficient wi (i = 1,..., M) of each evaluation item in the m evaluation items is determined. Here, an example will be described in which the weighting coefficient of each evaluation item is calculated using the following main eigenvector method (Equation 2). Here, for the paired comparison matrix A, L and q (q1, q2,..., Qm) satisfying the equation (Equation 2) are calculated. Note that L is an eigenvalue of the matrix A, and q is an eigenvector of the matrix A.

(Equation 2)
Aq = Lq (2)

For a square matrix A of m rows and m columns, there are a maximum of m eigenvalues L satisfying the formula (Equation 2), and the maximum value among them is Lmax (where Lmax is not 0). In the main eigenvector method, when the maximum eigenvalue Lmax is applied to the equation (Equation 2), each element of the matrix A can be regarded as the ratio of the element of the eigenvector to Lmax shown in (Equation 3).

(Equation 3)

Thus, the element of the eigenvector q with respect to the maximum eigenvalue Lmax of the paired comparison matrix A can be set as the weighting coefficient wi (i = 1,..., M) of each evaluation item. Therefore, the formula for calculating the weight vector m corresponding to the paired comparison matrix A shown in Table 2 is the formula shown in (Equation 4).

・・・（４）

(Equation 4)

... (4)

  As a result of the calculation of (Expression 4), the “safety” weight (W1) for the “counterfeit-proof printed matter with excellent overall performance” is 0.578, and the “convenience” weight (W2) is 0.295. The “social acceptability” weight (W3) is 0.077, and the “cost” weight (W4) is 0.05.

However, normalization was performed so that w1 + w2 + w3 + w4 = 1 and wi ≧ 0. The weighting factor wi is calculated by calculating the weighting factor wi of each evaluation item by taking the geometric mean in the row direction of the paired comparison matrix A and dividing the geometric mean of each row by the sum of the geometric mean. Or the arithmetic average of the reciprocals in the row direction of the paired comparison matrix A, the harmonic average of each row is calculated by the sum of the harmonic averages, and the weighting coefficient wi is calculated by dividing the harmonic average of each row by the sum of the harmonic averages. A calculation method and the like are known (see, for example, Non-Patent Document 3), and the calculation method of the weighting coefficient wi in the hierarchical analysis method is not limited to the main eigenvector method described in this example.

The weighting coefficient wi obtained from the paired comparison matrix is merely a comparison of the importance of the two evaluation items. It is possible that mistakes are inherent in the person performing the evaluation, and inconsistencies such as the order of importance being reversed may occur. Therefore, the consistency of the paired comparison matrix A is determined based on a numerical value represented by an expression shown in (Expression 5) called a degree of consistency (CI: Consistency Index).

(Equation 5)

Here, Lmax is the maximum eigenvalue of the paired comparison matrix A, and m is the number of elements that have undergone paired comparison (in this case, the number of items of evaluation items). If this threshold is set to 0.15 and this threshold is exceeded, the paired comparison value is reexamined. Note that the degree of matching of the paired comparison matrix A shown in Table 2 is calculated as CI = 0.111 from the above equation, and is determined to have sufficient matching.

  Next, the evaluation value in each evaluation item is input for a plurality of forgery prevention printed materials. For this evaluation, the degree of evaluation (pair comparison value) as shown in Table 3 is used.

(Table 3)

  Similarly to the procedure described in the pair comparison of the importance of the evaluation items, in each evaluation item, the paired comparison value when the forgery-preventing printed matter X is compared with the forgery-preventing printed matter Y is displayed in the column next to the forgery-preventing printed matter X (in the row direction). ). In this example, a description of the specific specifications and configuration of the anti-counterfeit printed matter X and the anti-counterfeit printed matter Y is omitted, but the paired comparison values shown in Table 4 indicate the values of the pairwise comparative evaluation. . On the other hand, in the vertical column (column direction) of the forgery prevention printed matter X, the reciprocal of the paired comparison value is input. In this example, forgery-proof printed matter X, anti-counterfeit printed matter Y, and anti-counterfeit printed matter Z are compared with each other in terms of safety, which is one of the evaluation items, between each forgery-proof printed matter. Table 4 shows an example of the result of the trial. As described above, the number of elements to be evaluated by paired comparison, that is, the number of anti-counterfeit printed matter to be evaluated is preferably within seven.

(Table 4)

Next, the partial evaluation raw value of each anti-counterfeit printed matter is calculated from a paired comparison value of a plurality of anti-counterfeit printed matters related to safety shown in Table 4. The partial evaluation raw value is a raw score for evaluation in an arbitrary evaluation item of each anti-counterfeit printed matter, and here, the partial evaluation raw values of the anti-counterfeit printed matter X, Y, and X related to “safety” will be described. . When the number of anti-counterfeit printed materials to be evaluated is n types, the partial evaluation values v1, k (k = 1,..., N) relating to “safety” of the anti-counterfeit printed materials are the weighting factors described above. Similar to the eigenvector calculation method shown in the calculation, each counterfeit-prevented printed material corresponds to each element of the eigenvector for the maximum eigenvalue Kmax of the n-by-n paired comparison matrix B created from the viewpoint of security. The partial evaluation elementary values v1,1, v1,2, v1,3 corresponding to the paired comparison matrix B shown in Table 4 are calculated by the equation (Equation 6).

(Equation 6)

As a result of the calculation of (Equation 6), the partial evaluation elementary value (V1, 1) of “forgery prevention printed matter X” for “safety” is 0.487, and the partial evaluation elementary value of “forgery prevention prevention printed matter Y” ( V1,2) is 0.078, and the partial evaluation elementary value (V1,3) is 0.435.

The degree of consistency is evaluated with respect to the partial evaluation elementary values v1, k (k = 1,..., N) of each forgery-preventing printed matter regarding “safety” as an evaluation item. The degree of matching of the partial evaluation elementary values v1, k (k = 1,..., N) in this example was 0.006, and it was determined that the matching was sufficient.

Similar to the calculation of the partial evaluation elementary values related to “safety”, the paired comparison values performed when calculating the partial evaluation elementary values v2, k (k = 1,..., N) related to “convenience” Table 5 shows the results of partial evaluation elementary values and matching degrees calculated by the paired comparison values.

(Table 5)

Also, partial evaluation values v3, k (k = 1,..., N) relating to “social acceptability”, and partial evaluations calculated using paired comparison values, which were performed when calculating the paired comparison values. Table 6 shows the results of the elementary values and the degree of matching.

(Table 6)

In addition, an example of a pair comparison value performed when calculating the partial evaluation elementary values v4, k (k = 1,..., N) regarding “cost”, a partial evaluation elementary value calculated by the pair comparison value, and Table 7 shows the results of the degree of matching.

(Table 7)

  Next, the partial evaluation elementary values v1, k, v2, k, v3, k, v4, k (k = 1,...) Regarding the four evaluation items obtained by the pairwise comparison between the forgery prevention printed materials. From n), the comprehensive evaluation value tk (k = 1,..., n) for each forgery-preventing printed matter is calculated based on the equation (Equation 7).

・・・（７）

(Equation 7)

... (7)

Here, each of w1v1, k, w2v2, k, w3v3, k, w4v4, k (k = 1,..., N) is an evaluation item in each forgery prevention printed matter “safety”, “convenience” ”,“ Social acceptability ”, and“ cost ”, which are values obtained by multiplying the partial evaluation elementary values and the weighting coefficient, are set as partial evaluation values.

Finally, Table 8 shows an evaluation list created based on the partial evaluation value and the comprehensive evaluation value of each evaluation item of each anti-counterfeit printed matter in this example. Similarly, FIG. 2 shows an evaluation profile created based on the evaluation list shown in Table 8. Here, FIG. 2A shows a comprehensive evaluation value of each printed matter as a bar graph, and FIG. 2B shows a partial evaluation value of each evaluation item as a radar chart.

(Table 8)

As described above, the comprehensive evaluation of the forgery prevention printed matter using the hierarchical analysis method by the present inventor and the example of visualization of the evaluation result have been described.

Japanese Patent No. 4549514 JP-A-9-69051

Eiichiro Takatsuki and Nobuyuki Nakajima, “Introduction to AHP with Excel,” Ohmsha, September 2005) Une, Tamura, Matsumoto, Artifact Metrics in Anti-Counterfeiting Technology: Trends and Issues in Security Research and Development, Financial Research, 2009, 7) Akihisa Hoshida, a new application of the hierarchical analysis method: Theory and Applications of GIS, 2006, Vol. No., focusing on the quantification of many aspects and the examination of the evaluation score matrix. 2, pp. 63-72)

Anti-counterfeiting prints such as banknotes and passports are broadly classified into safety requirements such as anti-counterfeiting resistance, convenience requirements such as usability and durability, social aspects such as aesthetics and suitability for social culture. It is required to satisfy the acceptability requirements and the cost requirements such as materials and manufacturing costs in a well-balanced manner. Of these, some of the physical and chemical durability of anti-counterfeit printed materials have been evaluated quantitatively, but these are evaluations for individual evaluation items, and different types of evaluation results are merged. It is not a comprehensive and comprehensive assessment cannot be performed. For anti-counterfeit printed matter such as banknotes and passports that play an important role in financial transactions and personal authentication situations, a comprehensive evaluation method with high accuracy, reliability and rationality is required.

  As described above, by using the hierarchical analysis method used in Patent Document 1 and Patent Document 2 for comprehensive evaluation of anti-counterfeit printed material, it is considered that rough general evaluation of anti-counterfeit printed material can be performed to a certain extent. However, when performing the comprehensive evaluation of the forgery-preventing printed matter using the hierarchical analysis method described above, there are the following problems. For example, when a printed material to be evaluated is added, there is a problem that it is necessary to perform a pair comparison between the printed materials to be evaluated again. On the other hand, if the technique of patent document 1 is used, the performance represented by the quantitative value can be evaluated without taking time and cost, and if the technique of patent document 2 is used, an alternative is added. In this case, the evaluation can be performed without making a pair comparison again. However, the evaluation methods of Patent Document 1 and Patent Document 2 have the following problems.

  The facility performance evaluation system disclosed in Patent Document 1 is intended to roughly evaluate performance without labor and cost, and determines a performance level for each evaluation item, and sets a multiple for each level. An evaluation score is obtained by multiplying the weighting factor of the item by a multiple of the level corresponding to the evaluation target. The setting of this performance level is an extremely simple setting method, and it is difficult to say that it is sufficient from the viewpoint of accuracy, rationality, and reliability of the evaluation result. For example, when setting the performance level, the time from the nearest station of the facility is set within the order of 1 minute, 10 minutes, and 20 minutes, and the multiples are simply set in the order of 3 times, 2 times, and 1 time. When comparing 10 minutes for a 20 minute facility and when evaluating a 1 minute facility for a 10 minute facility, it is not possible to evaluate with the same multiple setting. There is. In other words, facilities whose time is short from the nearest station should increase in value and value, but cannot be evaluated from the performance level of Patent Document 1.

  Moreover, in the apparatus which evaluates the alternative shown by patent document 2, there existed a problem that the method of handling a quantitative evaluation result was not enough. In the technique of Patent Document 2, the quantitative evaluation result is handled by dividing the quantitative value D (i), which is the quantitative evaluation result of the alternative in any evaluation item, by the maximum value Dmax of the quantitative evaluation result between the alternatives. The evaluation is performed by multiplying the value and the weight. However, the possible value of the quantitative value D (i) is assumed to be various cases such as a small number, a fraction, a negative number, and a multiplier, and is a uniform calculation method called division by the maximum value Dmax. It is unlikely that the accuracy of the results will be guaranteed. For example, there is a problem that the evaluation value varies greatly depending on the magnitude relationship with the maximum value in the quantitative value to be evaluated.

Another problem is the elimination of evaluation accuracy degradation due to the mutual dependency between evaluation items derived as a tree structure. In general, the hierarchical analysis method requires mutual independence between evaluation items, but it is difficult in practice, and this has an adverse effect on the evaluation result. The basic requirements (evaluation items) for anti-counterfeit printed matter will be explained with concrete examples of safety, convenience, social acceptability, and cost. Overall, the higher the cost, the higher the safety. Similarly, those with high convenience tend to have high social acceptability. There is a concern that a high-accuracy comprehensive evaluation result cannot be given to a printed material that is low in cost but high in safety and should be given a high overall evaluation. That is, there is a need for a correction processing procedure that takes into account the case where each evaluation item is not sufficiently independent.

As described above, when summarizing the problems of the present invention, in conducting a comprehensive evaluation of anti-counterfeit prints using a plurality of requirements required for anti-counterfeit prints as evaluation items, the quantitative evaluation results of each evaluation item are accurate. It is intended to provide a comprehensive evaluation system and method for anti-counterfeit printed matter having reliability, rationality and reliability.

The method for evaluating the overall performance of the anti-counterfeit printed material of the present invention uses a plurality of requirements in the anti-counterfeit printed material as evaluation items, and uses the numerical values measured from the anti-counterfeit printed material for the evaluation items, A tree structure setting process in which a final goal is set at the highest level, and a plurality of requirements for realizing the higher level target are set as evaluation items, and a tree structure setting process. A weighting factor calculating step of calculating a weighting factor of the evaluation item from the first eigenvector by pair-wise comparison between the evaluation items in the same hierarchy set in step 1, and a forgery-preventing printed matter to be evaluated in each of the set evaluation items Set multiple levels to classify numerical values measured from the second eigenvector by pairwise comparison between levels set in each evaluation item The performance weight coefficient for each level is calculated, and the performance weight coefficient set calculation process for one group, and for each set evaluation item, the numerical value measured from the anti-counterfeit printed matter to be evaluated is the performance weight coefficient. Classify into one of multiple levels set in the set calculation process, and divide the performance weight coefficient corresponding to the classified level by the maximum value of the performance weight coefficient that constitutes the same group to evaluate the partial performance Partial performance evaluation raw value calculation step for calculating a raw value, partial performance evaluation raw value calculated in the partial performance evaluation raw value calculation step, and a weighting coefficient corresponding to the partial performance evaluation raw value are multiplied by the partial performance evaluation value At least a partial performance evaluation value calculating step for calculating the partial performance evaluation value and a total performance evaluation value calculating step for performing a total calculation of the partial performance evaluation values calculated in the partial performance evaluation value calculating step.

In addition, the method for evaluating the overall performance of the forgery-preventing printed material according to the present invention is a pair for evaluating the degree of influence of each evaluation item on one evaluation item in the evaluation items in the same hierarchy set in the tree structure setting step. A weighting factor correction step is performed between the weighting factor calculation step and the partial performance evaluation value calculation step to calculate an interdependent weighting factor from the third eigenvector by comparison and to multiply the weighting factor by the interdependent weighting factor for correction. It is characterized by.

  In addition, the overall performance evaluation system for anti-counterfeit printed materials of the present invention uses a plurality of requirements for anti-counterfeit printed materials as evaluation items, and uses numerical values measured from the anti-counterfeit printed materials for the evaluation items. It is a system for evaluating performance, and a tree structure setting unit for setting a hierarchical structure with a plurality of requirements in an anti-counterfeit printed matter as evaluation items, and each evaluation item set in the tree structure setting unit, The numerical values measured from the anti-counterfeit printed material, the multiple levels for classifying the measured numerical values, and the evaluation items in the same hierarchy set by the levels and the tree structure setting unit, for each pair comparison The weight coefficient is determined from the eigenvector of each paired comparison value input at the input unit and at least the paired comparison value to be used. For each evaluation item, classify the numerical value measured from the anti-counterfeit printed matter to be evaluated, which is input at the input section, into one of multiple levels, and assign a weighting factor corresponding to the classified level. Divide by the maximum value of the weighting factor at multiple levels of the same evaluation item, multiply the divided value by the weighting factor calculated for each evaluation item, and calculate the evaluation value for each evaluation item. It has a control part provided with the evaluation processing part which adds up an evaluation value and performs comprehensive evaluation.

  Further, in the overall performance evaluation system for forgery-preventing printed matter of the present invention, the tree structure setting unit has at least the highest order as the final purpose and a plurality of evaluation items for realizing the high order purpose below. It comprises a hierarchy setting section for setting two or more hierarchical structures, and an evaluation item input section for setting evaluation items in the set hierarchical structures.

  Further, in the comprehensive performance evaluation system for forgery-preventing printed matter of the present invention, the data input unit is a performance level input unit for inputting values of intervals of a plurality of levels, and a pair used for pairwise comparison between levels and evaluation items. A weight input unit for inputting a comparison value and an evaluation value input unit for inputting a numerical value measured from an anti-counterfeit printed matter to be evaluated are characterized.

  Further, in the overall performance evaluation system for forgery-preventing printed matter of the present invention, the evaluation processing unit calculates a weighting factor of a plurality of levels for each evaluation item, a weighting factor calculation unit that calculates a weighting factor of each evaluation item, Performance weight coefficient set calculation unit for one group, and for each evaluation item, the numerical values measured from the forgery-proof printed matter to be evaluated are classified into one of multiple levels to form the same group Each partial performance evaluation value calculation unit that divides by the maximum value of the weighting factor, each evaluation item that is calculated by the partial performance evaluation value calculation unit that multiplies the divided value and the weighting factor of each evaluation item, and each partial performance evaluation value calculation unit And a total performance evaluation value calculation unit for adding the evaluation values.

  In the comprehensive performance evaluation system for forgery-preventing printed matter according to the present invention, the data input unit is further used for pairwise comparison for evaluating the degree of influence of each evaluation item on one evaluation item in the evaluation items in the same hierarchy. The paired comparison value is input, and the evaluation processing unit is based on the paired comparison value that evaluates the degree of influence of each evaluation item on one evaluation item in the evaluation items in the same hierarchy input by the data input unit. A weighting factor is calculated from the eigenvector, and the weighting factor calculated for each evaluation item is multiplied and corrected.

  Further, in the overall performance evaluation system for forgery-preventing printed matter of the present invention, the data input unit uses a pair comparison for evaluating a degree of influence of each evaluation item on one evaluation item in the evaluation items in the same hierarchy. The evaluation processing unit includes a dependency coefficient calculation unit that calculates a weighting coefficient from an eigenvector based on a pair of comparison values input by the mutual dependency value input unit, and an evaluation item. It is characterized by having a weight correction calculation unit that corrects by multiplying the calculated weight coefficient by the weight coefficient calculated by the dependency coefficient calculation unit.

  Further, the total performance evaluation system for anti-counterfeit printed matter of the present invention includes a list of comprehensive evaluation values obtained by adding together the evaluation value of each evaluation item and the evaluation value of each evaluation item in the anti-counterfeit printed matter to be evaluated. Alternatively, it further includes an evaluation output unit that generates and outputs the data in a predetermined format.

Further, in the comprehensive performance evaluation system for forgery-preventing printed matter of the present invention, the evaluation output unit creates an evaluation list for each evaluation item and a list of values for comprehensive evaluation, and outputs an evaluation list creation unit for each evaluation item. It is characterized by comprising an evaluation profile creation unit that creates and outputs an evaluation value and a comprehensive evaluation value in a predetermined format.

According to the present invention, in each evaluation item set for comprehensive evaluation of the anti-counterfeit printed matter, quantitative values measured from the anti-counterfeit printed matter are assigned to a plurality of levels, and weighted by a pairwise comparison between the levels. By evaluating later, the partial performance evaluation value corresponding to each evaluation item and the comprehensive evaluation value of the anti-counterfeit printed matter to be evaluated can be accurately and reasonably obtained.

Moreover, according to this invention, in addition to the improvement of the evaluation precision by the setting of a performance level, the evaluation precision can be improved more by using the weighting coefficient which evaluated the influence degree of evaluation items.

It is a figure which shows an example of the evaluation tree structure for achieving the forgery prevention printed matter excellent in total performance. It is a figure which shows the total evaluation value in the forgery prevention printed matter, and the partial evaluation value of total performance. It is a block diagram which shows the structure of the comprehensive evaluation system of the forgery prevention printed matter of this invention. It is a flowchart of the method of evaluating the comprehensive performance of the forgery prevention printed matter of this invention. It is a figure which shows the detail of the setting process of an evaluation tree structure. It is a figure which shows an example of the setting of a tree structure. It is a figure which shows the detail of the calculation process of the weighting coefficient between each evaluation item. It is a figure which shows the detail of the calculation process of a performance weighting coefficient set. It is a figure which shows the detail of the calculation process of a partial performance evaluation elementary value. It is a figure which shows the detail of the calculation process of a partial performance evaluation value. It is a figure which shows the detail of the calculation process of a comprehensive performance evaluation value. It is a figure which shows an example of visualization of the comprehensive performance evaluation value regarding the discriminability of each printed matter, and a partial performance evaluation value. It is a block diagram which shows the structure of the forgery prevention printed matter comprehensive evaluation system of 2nd Embodiment. It is a flowchart of the method of evaluating the comprehensive performance of the forgery prevention printed matter of 2nd Embodiment. It is a figure which shows the detail of the correction process of a weighting coefficient. It is a figure which shows an example of visualization of the comprehensive performance evaluation value regarding the discriminability of each printed matter, and a partial performance evaluation value in consideration of the interdependency between each evaluation item.

Embodiments of the present invention will be described in detail with reference to the drawings. However, the present invention is not limited to the embodiments described below, and various other embodiments can be implemented within the scope of the technical idea described in the claims.

(First embodiment)
(Comprehensive evaluation system for anti-counterfeit printed matter)
FIG. 3 is a block diagram showing the configuration of the forgery-proof printed matter comprehensive evaluation system (A) of the present invention. The anti-counterfeit printed overall evaluation system (A) has a control unit (C) as a core and is roughly divided into an evaluation tree structure setting unit (S), a data input unit (I), and an evaluation output unit (O). Although specific functions of each part will be described later, a series of functions can be implemented using spreadsheet software, numerical analysis software, or the like that operates on a general-purpose personal computer.

(Evaluation tree structure setting part)
The evaluation tree structure setting unit (S) sets the mutual relationship between the evaluation items for evaluation derived when performing the comprehensive evaluation of the forgery prevention printed matter, that is, the tree structure and the specific item names of the evaluation items. It is a part.

As shown in FIG. 1, the evaluation tree structure setting unit (S) has a hierarchical setting unit (set at least two or more hierarchical structures including a plurality of evaluation items for realizing the purpose, with the highest order as the final purpose. S1) and an evaluation item input unit (S2) for setting specific evaluation items in the hierarchical structure. The hierarchy setting unit (S1) uses a hierarchical structure drawing component that can be generally used by a word processor or drawing software to generate a character string input box or a function that arbitrarily connects character string input boxes with ruled lines. Has a function of determining a subordinate (hierarchical) relationship of blank character string input boxes into which a plurality of evaluation items are input, and setting an arbitrary tree structure. The function of setting the hierarchy can be replaced by a spreadsheet software cell sheet and ruled line function. The evaluation item input unit (S2) is an arbitrary tree structure composed of character string input boxes or cells set by the hierarchy setting unit (S1), and is activated by the input box or cell activation by mouse instruction or cursor movement. It has an input function for final purpose or specific evaluation items. Note that the set and entered hierarchical structure and evaluation items are stored in the storage unit (M) in the form of graphic information or cell sheets, and can be freely read, edited, and stored.

(Data input part)
The data input unit (I) includes a weight input unit (I1), a performance level value input unit (I2), and an evaluation value input unit (I3). The weight input unit (I1) is the importance or priority when the paired comparison is performed between the evaluation items set in the same hierarchy in the evaluation items of the hierarchical structure set to realize the final purpose, that is, This is the part to input the weight. In addition, as shown in Table 1 and Table 3, the input of weights defines the degree of importance when paired comparisons are made between evaluation items of the same hierarchy for higher-level objectives, and the importance of the defined importance is defined. A paired comparison value set for each degree is used. The weight input unit (I1) also inputs a paired comparison value that is a degree of importance or priority when a pairwise comparison is made between performance levels set by the performance level value input unit (I2) to be described later. . The paired comparison values shown in Table 1 and Table 3 input by the weight input unit (I1) may be stored in advance in the storage unit (M), read out, and the corresponding value may be input. Then, at the stage of inputting the weight for the evaluation item, a paired comparison value may be appropriately defined and input. Further, in order to alleviate the variation of the paired comparison values when a plurality of evaluators perform the paired comparison, a part for taking a geometric average of the paired comparison values input by the weight input unit may be provided (not shown). .

The performance level input unit (I2) is for classifying the numerical performance of the forgery-preventing printed matter for the evaluation items (for example, quantitative evaluation results such as time required for authenticity determination and minimum required illuminance). This is a part for inputting a section for setting a plurality of levels. Hereinafter, the level of multiple levels for the evaluation item is referred to as the performance level. The first feature of the present invention is that the performance level value input unit (I2) of the previous operation is provided. For example, the performance level value input unit (I2) sets the levels shown in Table 9 for evaluation items such as the time required for authenticity determination of a forgery-preventing printed material. In addition, the level of the identification time shown in Table 9 is set as four stages of less than 0.6 seconds, 0.6 seconds to 2 seconds, 2.1 seconds to less than 3.5 seconds, and 3.5 seconds or more. However, in the present invention, the numerical range of the set section and the number of levels are not limited to this.

(Table 9)

The evaluation value input unit (I3) is a part that inputs an evaluation result given as a quantitative value in each evaluation item. For example, when the identification time is set as one of the evaluation items, the time required for the identification in the forgery prevention printed matter to be evaluated is input using the evaluation value input unit (I3).

(Control part)
The control unit is roughly divided into an evaluation processing unit (P) and a storage unit (M). The evaluation processing unit (P) uses the tree structure evaluation items set by the evaluation tree structure setting unit (S) and the data input by the data input unit (I) to evaluate the forgery prevention printed matter. A weighting factor calculation unit (P1), a matching degree calculation determination unit (P2), a performance weighting factor set calculation unit (P3), a partial performance evaluation elementary value calculation unit (P4), and a partial performance evaluation value calculation unit (P5). And an overall performance evaluation value calculation unit (P6). The storage unit (M) connected to the evaluation processing unit (P) includes tree-structured evaluation items, weight coefficients, consistency, partial performance evaluation elementary values, partial performance evaluation values, total performance evaluation values, and performance weights. Coefficient set values are stored.

The weighting factor calculation unit (P1) is a part that calculates the weighting factor of the evaluation items set in the same hierarchy from the paired comparison result input by the weight input unit (I1). It may be calculated by the main eigenvector method of the equations (Equation 1), (Equation 2), (Equation 3), and (Equation 4), or may be calculated by the geometric average method or the harmonic average method described above.

  The degree-of-matching calculation determination unit (P2) is a part that determines the degree of matching of the weighting factor calculated by the weighting factor calculation unit (P1), and the calculation method is calculated by the equation (Equation 5) described above. Can do.

  The performance weight coefficient set calculation unit (P3) is a part that is input by the performance level value input unit (I2) and performs calculation of the performance weight coefficient set from the input result of the pair comparison value by the pair comparison between the set levels. . The calculation method of the performance weight coefficient set may be calculated by the main eigenvector method of the expressions (Equation 1), (Equation 2), (Equation 3) and (Equation 4) described above, It may be calculated by a harmonic average method. These arithmetic methods can be implemented using general spreadsheet software or numerical analysis software.

  Here, the performance weight coefficient set is a group of weight coefficients for each level calculated when a pair of performance levels set for a specific evaluation item are compared. For example, in the case of the performance level shown in Table 9, four levels are set for the evaluation item of the identification time, and the four level weight coefficients calculated by the four level pair comparison are made into one group. . In the following description, in order to distinguish from the weighting coefficient of the evaluation item, the weighting coefficient of each level set as the performance level will be described as the performance weighting coefficient. As described above in the problem of the prior art, Patent Document 1 is an evaluation by simply multiplying each performance level, and Patent Document 2 is a uniform method of dividing the maximum value. Therefore, it is difficult to say that this is a highly accurate evaluation method. On the other hand, in the comprehensive evaluation system for anti-counterfeit printed material according to the present invention, the performance weight coefficient set calculation unit (P3) uses the performance weight coefficient set calculated through the pairwise comparison between the levels, so The accuracy of evaluation is remarkably improved by considering the weight (importance).

  The partial performance evaluation element value calculation unit (P4) is a partial performance evaluation element from the evaluation value input by the evaluation value input unit (I3) and the performance weight coefficient set calculated by the performance weight coefficient set calculation unit (P3). This is the part where the value is calculated. Here, the partial performance evaluation elementary value is obtained by dividing the performance weight coefficient corresponding to the level to which the numerical performance of the evaluation target in a specific evaluation item belongs by the maximum value of the performance weight coefficient constituting the performance weight coefficient set. is there. For example, if the performance level shown in Table 9 is set and the identification time of the anti-counterfeit printed matter A to be evaluated is 3 seconds, the weighting factor of the level from 2.1 seconds to 3.5 seconds is set as the performance weighting factor. Reference is made from the set, and division is performed by the maximum value of the weight coefficients of the four levels constituting the performance weight coefficient set of the evaluation item of the identification time. Although details of the calculation of the partial performance evaluation elementary values will be described later, this calculation method can be implemented using general spreadsheet software, numerical analysis software, or the like.

  The partial performance evaluation value calculation unit (P5) uses the weighting coefficient calculated by the weighting coefficient calculation unit (P1) and the partial performance evaluation element value calculated by the partial performance evaluation element value calculation unit (P4) for each evaluation item. This is a part where the partial performance evaluation value is calculated by multiplying. This calculation can be performed using general spreadsheet software or numerical analysis software.

  The total performance evaluation value calculation unit (P6) is a part that calculates the total performance evaluation value by adding the partial performance evaluation values of the respective evaluation items calculated by the partial performance evaluation value calculation unit (P5). This calculation is also implemented using spreadsheet software.

(Evaluation output part)
The evaluation output unit (O) creates and outputs a list composed of the name of the anti-counterfeit printed matter to be evaluated, each evaluation item, the evaluation value of each evaluation item, and the overall evaluation value, and outputs the evaluation list (O1). An evaluation profile creating unit (O2) that creates an evaluation profile in which evaluation values corresponding to each evaluation item are visualized in a radar chart format and outputs the result on a paper surface. The output of the comprehensive evaluation result is not limited to a paper surface using a general-purpose printer connected to a personal computer, and can be replaced by generating an electronic document file such as PDF.

Various calculations related to the comprehensive evaluation of the forgery-preventing printed matter performed by the control unit (C) are executed by a CPU of a general-purpose personal computer constituting the system. The storage unit (M) that stores the calculation results in the CPU and reads out the numerical files necessary for various calculations is a hard disk of a general-purpose personal computer or an arbitrary recording device on the cloud.

Next, the comprehensive evaluation method for forgery-preventing printed matter of the present invention using the forgery-preventing printed matter comprehensive evaluation system (A) will be described. FIG. 4 is a flowchart showing an outline of a comprehensive evaluation method for forgery-preventing printed matter. Details of the evaluation tree structure setting step (F1) of FIG. 4 are shown in FIG. The first step (F1-1) in FIG. 5 is setting of a tree structure using the hierarchy setting unit (S1). FIG. 6A is an example of the set tree structure. An input box for inputting a final target is arranged at the vertex, and an input box for inputting a plurality of large evaluation items is arranged below the large evaluation. An input box for inputting a plurality of middle evaluation items is arranged below the item. These input boxes are preferably connected by ruled lines to facilitate understanding of the dependency relationship.

  Next, the second step (F1-2) in FIG. 5 is a setting of the final purpose located at the top of the evaluation tree. Here, as shown in FIG. "Anti-counterfeit printed matter" is set as the final purpose. The third step (F1-3) in FIG. 5 is setting of evaluation items. As an example of evaluation items necessary for comprehensive evaluation in the forgery prevention printed matter, as shown in FIG. 6B, “safety”, “convenience”, and “social acceptability” are set in the evaluation item input unit ( Using S2), settings are made in a form input to a hierarchical structure drawing component of a word processor or drawing software or a spreadsheet sheet cell.

Note that the number of large evaluation items in the tree structure shown in FIG. 6B is merely an example of the embodiment of the present invention, and three evaluation items are set. The number of evaluation items to be performed is not limited to this. The number of evaluation items set in the same hierarchy is the same as this. However, as mentioned above, if the number of objects to be compared by paired comparison is more than seven, it becomes difficult to ensure consistency when evaluating each evaluation item with an overall scale. It is desirable that the number of evaluation items set within is 7 or less.

  FIG. 6B is an example in which a large evaluation item is set as a tree structure. However, when the content of the large evaluation item is further subdivided and detailed evaluation is desired, A plurality of middle evaluation items may be set. For example, medium evaluation items such as authenticity, completeness, and confidentiality can be set for the safety set as the large evaluation items shown in FIG. In this way, detailed evaluation can be performed by setting more specific evaluation items for evaluating higher evaluation items.

The fourth step (F1-4) in FIG. 5 is to store the evaluation tree structure, and the set tree structure is stored in the storage unit (M). Note that the evaluation tree structure setting step (F1) in FIG. 4 may be a read correction, storage, and reference function in addition to the new creation of the tree structure described so far. Here, an example of setting the final purpose and the evaluation items for the purpose in the evaluation tree structure setting step (F1) will be described.

As described above, the present invention performs a comprehensive evaluation of the anti-counterfeit printed matter. Here, in order to briefly explain the evaluation method of the present invention, as shown in FIG. Take an example in which “identity”, one of the requirements, is set as the final goal. Using the evaluation item input unit (S2) in the evaluation tree structure setting unit (S), “an anti-counterfeit printed matter having excellent identification” was input as the final purpose. Also, following the details of the flowchart shown in FIG. 5, a tree structure of evaluation for achieving “an anti-counterfeit printed matter with excellent identification” is set, and a subordinate specific to “an anti-counterfeit printed matter with excellent identification” Identification time, expression angle, and identification illuminance were input as typical evaluation items and stored in the storage unit (M). Here, the identification time is a time required for authenticity determination. In addition, the expression angle is an angle at which a latent image appears. For example, the printed matter described in Japanese Patent No. 2615401 cannot be visually recognized when viewed from the front, but from the front when viewed obliquely. An image that cannot be viewed is visible as a latent image. Other techniques for visually recognizing latent images include holograms and optically changing inks. Further, the identification illuminance is the brightness that is at least required for authenticity determination. The result of the evaluation tree structure set by the evaluation tree structure setting step (F1) in FIG. 4 is shown in FIG.

In the present embodiment, an example will be described in which “identification time”, “expression angle”, and “identification illuminance” are set as lower specific evaluation items for “counterfeit prevention printed matter with excellent identification”. The setting of specific evaluation items is not limited to this, and any evaluation item may be set as necessary. For example, the “environmental impact” evaluation item is selected if there are items for which the emission standards for pollutants are stipulated by various laws and regulations, while the “durability” evaluation item includes ISO and JIS. It is required to cover the evaluation methods specified in the standard document. In addition, from the viewpoint of the user to consider “convenience” such as usability, and from the viewpoint of the producer and product orderer to consider “cost” such as raw material costs and manufacturing suitability, Evaluation items that achieve the final purpose may be derived.

As indicated by the broken lines in FIG. 6, a middle evaluation item that is a lower layer of the large evaluation item, and a small evaluation item (not shown) that is a lower layer of the middle evaluation item are used as necessary. Can be set. Similarly, it is possible to assume that “an anti-counterfeit printed matter having excellent identification” is a sub-set of a higher hierarchy, that is, one of the large evaluation items.

Note that the tree structure setting method described in this embodiment is to set the basic requirements as high-level evaluation items and then set the specific low-level evaluation items. However, it is not limited to this method. For multiple specific evaluation items of anti-counterfeit printed material, evaluation items of similar concepts are grouped, and the upper concepts of the grouped evaluation items are set as a higher evaluation item in a tree structure. It can also take the form. For example, if there is a series of stress load tests such as wear, bending, pulling, impact, and pressure as an evaluation test for anti-counterfeit printed matter, the concept that can include these evaluation items, that is, "durability" is a superordinate concept. Can be derived and set as part of the evaluation tree structure.

Next, FIG. 7 shows details of the weighting factor calculation step (F2) shown in FIG. The first step (F2-1) in FIG. 7 is a paired comparison value input step. Here, as in the above-described method, a paired comparison is performed regarding the importance of each evaluation item (which can also be called a priority), and a paired comparison value is input using the weight input unit (I1) in FIG. . As an example, Table 9 shows the results of performing weight input of identification time, expression angle, and identification illuminance, which are lower evaluation items related to discrimination.

(Table 9)

In the second step (F2-2) in FIG. 7, the weighting factors wi (i = 1,..., M) for each evaluation item are calculated. The calculation method of the weighting coefficient is based on the main eigenvector method shown in (Equation 2), (Equation 3), and (Equation 4) described above, and uses the weighting coefficient operation unit (P1) in the control unit (C) in FIG. Do. As an example, Table 10 shows the calculation result of the weight coefficient Wi of the paired comparison matrix described in Table 9 based on the main eigenvector method. Note that, as described above, other methods using a geometric average or a harmonic average can be applied to the calculation of the weighting factor.

(Table 10)

The third step (F2-3) in FIG. 7 is a calculation of the degree of matching of the calculated weighting factors. The calculation method is calculated by the matching degree calculation determination unit (P2) in the control unit (C) of FIG. 3 based on the equation (Equation 5) described above. The degree of matching of the weighting factors shown in Table 10 was calculated to be 0.014.

The fourth step (F2-4) in FIG. 7 is determination of the calculated degree of matching. If the degree of matching calculated by the degree-of-matching determination unit (P2) exceeds 0.15, there is a concern that the weighting includes blurring and fluctuations, so the first step (F2- Return to 1) and re-enter the weights. In this specific example, the matching degree = 0.014 is determined to be appropriate by the matching degree calculation determination unit (P2) in the control unit (C) of FIG. Finally, the fifth step (F2-5) in FIG. The weighting factors in this example shown in Table 10 are stored in the storage unit (M) in FIG.

  In the performance weight coefficient set calculation step (F3) in FIG. 4, a performance weight coefficient set Pi (i = 1, 2, 3,..., N) is calculated. FIG. 8 shows a detailed flow of the calculation step (F3) of the performance weight coefficient set Pi (i = 1, 2, 3,..., N) in FIG. The first step (F3-1) in FIG. 8 is setting the performance level. Here, a method for setting the performance level of the identification time will be described. As described above, the performance level is a level composed of a plurality of stages in each evaluation item for which the evaluation result of the anti-counterfeit printed matter is given by numerical performance. For example, in the case of the identification time, the short level is less than 1 second, the long level is 5 seconds or more, and the intermediate level is 1 second or more and less than 5 seconds. Such a section is set using the performance level input unit (I2). Table 11 shows an example of setting the performance level of identification time, which is an evaluation item for identification.

(Table 11)

In this example, the performance level of banknote identification time was set in four stages as shown in Table 11. These values are based on an estimate that the average grace period given to cash handlers such as retailers and banks is 0.6 (seconds), and the average change confirmation time is about 2 (seconds). . Note that the setting of the performance level is not limited to the four levels in Table 11, and the setting of the level section is not limited to the example shown in Table 11. It may be set in consideration of the user's point of view, the producer's point of view, and the product orderer's point of view, or the maximum identification time among a plurality of evaluation objects may be equally divided and set. In consideration of a pair comparison process described later, the number of levels to be set is preferably within seven. This is because, as described above, if the number of objects to be compared is more than seven by pairwise comparison, it becomes difficult to ensure consistency when evaluating each evaluation item with an overall scale. .

The second step (F3-2) in FIG. 8 is a pairwise comparison between performance levels. This compares the superiority and inferiority of each performance level set in each evaluation item by the pair comparison method described above, and inputs the pair comparison value using the weight input unit (I1) shown in FIG. As an example, Table 12 shows paired comparison results between the four performance levels of the identification time shown in Table 11.

(Table 12)

The third step (F3-3) in FIG. 8 is calculation of the performance weight coefficient set Pi (i = 1, 2, 3,..., N). Based on the paired comparison values shown in Table 12, the performance weight coefficient set is based on the main eigenvector method shown in (Equation 2), (Equation 3), and (Equation 4) described above. It is calculated using the performance weight coefficient set calculation unit (P3). As an example, Table 13 shows the calculation result of the performance weight coefficient set Pi (i = 4) from the paired comparison results of the identification times shown in Table 12.

(Table 13)
Calculation result of performance weight coefficient set for identification time

The fourth step (F3-4) in FIG. 8 is calculation of the degree of matching of the calculated performance weight coefficient set. Based on the equation (Equation 5) described above, the calculation is performed using the matching degree calculation determination unit (P2) in the control unit (C). The matching degree of the performance weight coefficient set Pi (i = 4) shown in Table 13 was calculated as 0.038.

The fifth step (F3-5) in FIG. 8 is determination of the calculated degree of matching. If the degree of matching calculated in the fourth step (F3-4) exceeds 0.15, the pairwise comparison between the performance levels is likely to include blurring and fluctuations. Returning to step 2 (F3-2), the paired comparison value is re-input. The degree of matching = 0.038 in this example was determined to be appropriate by the degree of matching calculation determination unit (P2) in the control unit (C). Finally, the sixth step (F3-6) in FIG. 8 is saving the performance weight coefficient set. The performance weight coefficient set Pi (i = 4) shown in Table 13 is stored in the storage unit (M) in the control unit (C) after the matching degree determination.

It should be noted that other evaluation items for ensuring discrimination, such as the setting of the performance level in the expression angle and identification illuminance, the performance weight coefficient set Pi (expression angle: i = 5, identification illuminance: i = 4), and the degree of consistency Table 14 shows an example of the calculation result. The performance weight coefficient set for the expression angle and the identification illuminance is also stored in the storage unit (M) in the control unit (C) after the matching degree is determined.

(Table 14)

Next, in the partial performance evaluation elementary value calculation step (F4) in FIG. 4, a partial performance evaluation elementary value is calculated. In the present invention, the partial performance evaluation elementary value indicates an evaluation value of a requirement item or an evaluation item for achieving an arbitrary target positioned higher in the evaluation tree structure. However, the partial performance evaluation raw values do not take into account weights that mean the importance (or priority) of each evaluation item. FIG. 9 shows a detailed flow for calculating the partial performance evaluation raw value. The first step (F4-1) in FIG. 9 is reading of the performance weight coefficient set. Here, the performance weight coefficient set Pi (i = 1, 2, 3,...) Calculated in the performance weight coefficient set calculation step shown in FIG. 8 and stored in the storage unit (M) in the control unit (C). n) is read.

In the second step (F4-2) in FIG. 9, the corresponding performance weight coefficient is input. The corresponding performance weight coefficient is a performance weight coefficient corresponding to the performance level to which the evaluation result given by the quantitative value belongs in the performance weight coefficient set Pi. This is performed using the evaluation value input unit (I3). As an example of performance weight coefficient input using the evaluation value input unit (I3), a performance weight coefficient set is read. For example, in the case of identification time, the performance weight coefficient set shown in Table 13 is read and given as a quantitative value. If the operator manually inputs the performance weight coefficient corresponding to the performance level to which the evaluation result belongs, or if it is within 0.6 (seconds) using a logical function that can be used in a general spreadsheet sheet, for example, an IF statement It is also possible to use a function in which, after manually inputting an evaluation result, such as substituting 0.565, a performance weight coefficient corresponding to the input evaluation result is automatically input as a condition.

The third step (F4-3) in FIG. 9 is a division by the maximum value of the performance weight coefficient set of the performance weight coefficient input in the second step (F4-2). That is, when the maximum value of the read performance weight coefficient set is Pmax and the performance weight coefficient Pi corresponding to the performance level to which the evaluation result given as the quantitative value belongs is Pi, the partial performance evaluation elementary value V is expressed by (Equation 8). It is given by the division shown in the equation. For example, in the case of the performance weight coefficient set for the identification time shown in Table 13, the maximum value of the performance weight coefficient set is 0.565, and is input in the second step (F4-2) with the maximum value of the performance weight coefficient set. The obtained performance weight coefficient Pi is divided. This calculation is performed in the partial performance evaluation elementary value calculation unit (P5) in the control unit (C).

・・・（８） (Equation 8)

... (8)

Finally, the 4th process (F4-4) in Drawing 9 is preservation of partial performance evaluation elementary value V computed at the 3rd process (F4-3). The partial performance evaluation elementary value V is stored in the storage unit (M) in the control unit (C).

For the partial performance evaluation elementary value calculation step (F4) in FIG. 4, as an example, a description of a specific method for calculating the partial performance evaluation elementary values in the evaluation of the identification time, the expression angle, and the identification illuminance, which are evaluations of discrimination Do. Table 15 shows the evaluation results regarding the identification time, the angle, and the illuminance given in advance as quantitative values to the forgery-preventing printed materials S, T, and U. The identification time, angle, and illuminance values shown in Table 15 indicate provisional evaluation values for the provisional anti-counterfeit printed matter. When actually evaluating, banknotes and What is necessary is just to input the evaluation value in forgery prevention printed matter, such as various securities.

(Table 15)

As the first step (F4-1) in FIG. 9, a performance weight coefficient set (Table 13 and Table 14) corresponding to the identification time, the expression angle, and the identification illuminance is stored from the storage unit (M) in the control unit (C). Next, with reference to the read performance weight coefficient set, the second step (F4-2) in FIG. 9 corresponds to the performance level and performance level of the section corresponding to each evaluation result shown in Table 15. The performance weight coefficient Pi and the maximum value Pmax of the performance weight coefficient were input as shown in Table 16. In addition, it is also possible to use a maximum value function that can be used in a general spreadsheet sheet so that the maximum value is automatically input.

(Table 16)

Next, as the third step (F4-3) in FIG. 9, the performance weight coefficient Pi is divided by the maximum value Pmax of the performance weight coefficient input in the second step (F4-2) in FIG. The partial performance evaluation elementary value is calculated, and the partial performance evaluation elementary value is stored in the storage unit (M) of the control unit (C) as the fourth step (F4-4) in FIG. Table 17 shows the calculation results of the partial performance evaluation elementary values V at the identification time, the observation angle, and the illuminance, which are evaluation items of each forgery prevention printed matter shown in Table 16.

(Table 17)

Next, a partial performance evaluation value R is calculated in the partial performance evaluation value calculation step (F5) in FIG. In the present invention, the partial performance evaluation value R is the importance of each evaluation item, that is, the evaluation value of each evaluation item in which the weighting factor is considered. The partial performance evaluation value R is a partial performance evaluation elementary value. And the weighting factor. This calculation is calculated in the partial performance evaluation value calculation unit (P5) in the control unit (C).

A detailed flow for calculating the partial performance evaluation value R is shown in FIG. The first step (F5-1) in FIG. 10 is the reading of the partial performance evaluation elementary value V from the storage unit (M) of the control unit (C). Here, the partial performance evaluation shown in Table 17 is performed. Reading of the prime value V is performed.

The second step (F5-2) in FIG. 10 is the reading of the weighting factor calculated in the weighting factor calculation step (F2) and stored in the storage unit (M). Here, the weighting factor shown in Table 10 is used. Is read out. The third step (F5-3) in FIG. 10 is a multiplication of a partial performance evaluation elementary value and a corresponding weighting factor performed using the partial performance evaluation value calculation unit (P5) of the control unit (C). Here, the read weighting factor of Table 10 and the partial performance evaluation value V of Table 17 are used, and the partial performance evaluation elementary values of the anti-counterfeit printed matter S, T, U of the same evaluation items are multiplied by the weighting factor. For example, the partial performance evaluation value R in the identification time of the forgery-preventing printed matter S shown in Table 18 is obtained by the calculation shown in the following (Equation 9).

(Equation 9) Partial performance evaluation value in identification time of anti-counterfeit printed matter S = 0.481 (weighting factor of identification time) × 0.098 (Partial performance evaluation elementary value in identification time of anti-counterfeit printed matter S) = 0.047

In the same manner as in (Equation 9), the partial performance evaluation values of the identification time, the expression angle, and the identification illuminance in each forgery prevention printed matter are calculated by multiplying the partial performance evaluation elementary value R shown in Table 18 by the corresponding weighting coefficient. Is done.

In the fourth step (F5-4) in FIG. 10, the calculated partial performance evaluation value R is stored, and the calculation result of the partial performance evaluation value R is stored in the storage unit (M) of the control unit (C). The

Next, the comprehensive performance evaluation value calculation step (F6) in FIG. In the present invention, the total performance evaluation value T is the sum of partial performance evaluation values R calculated for each evaluation item in each forgery prevention printed matter. The total performance evaluation value T is calculated in the total performance evaluation value calculation unit (P6) in the control unit (C).

FIG. 11 shows a detailed flow for calculating the total performance evaluation value. The first step (F6-1) in FIG. 11 is reading of the partial performance evaluation value (R) from the storage unit (M) of the control unit (C). Table 18 shows the partial performance evaluation values for each evaluation item.

(Table 18)

The second step (F6-2) in FIG. 11 is the summation of the partial performance evaluation values of each forgery-preventing printed matter performed using the total performance evaluation value calculation unit (P6) of the control unit (C). Table 19 shows the calculation result of the comprehensive performance evaluation value in each forgery prevention technique shown in Table 18.

(Table 19)

The third step (F6-3) in FIG. 11 is the preservation of the overall performance evaluation value T. The overall performance evaluation value T calculated in the second step (F6-2) is stored in the storage unit (M) of the control unit (C).

The output step (F7) in FIG. 4 is an output of the calculation result. Using the evaluation output unit (O) of FIG. 3, the partial performance evaluation value P calculated in the partial performance evaluation value calculation step (F5) and / or the total performance evaluation value calculated in the total performance evaluation value calculation step (F6). The result of the operation is output. FIG. 12A is a diagram showing an overall performance evaluation value by a bar graph as an example, and FIG. 12B is a diagram showing a partial performance evaluation value of each evaluation item by a radar chart. Note that the types of graphs are not limited to the example shown in FIG. 12, and various types of visualization methods are assumed.

The embodiment of the comprehensive evaluation system (A) and method for forgery prevention printed matter according to the present invention has been described above. The present invention is not limited to a single tree structure drawn with a solid line shown in FIG. When the identification time indicated as the large evaluation item in FIG. 6D has a lower tree structure (medium evaluation item), the sum of partial performance evaluation values of the middle evaluation item is calculated in advance, and the identification time Can be assumed (operated) as a partial performance evaluation value. Similarly, it can be assumed that the “counterfeit prevention printed matter with excellent identification” set as the final purpose in the example of the embodiment is a lower evaluation item subordinate to some higher evaluation item. For example, when “an anti-counterfeit printed matter with excellent identification” is one of the evaluation items for realizing “convenience”, an “anti-counterfeit printed matter with excellent identification” calculated in the present embodiment It can be assumed that the overall performance evaluation value is a partial performance evaluation value for evaluating “convenience”. In addition, when multiple evaluators perform pairwise comparisons between evaluation items and performance levels, a geometric mean value is calculated to reduce the variation of the paired comparison values by the evaluators, and the weight of each evaluation item is calculated. A coefficient and a performance weight coefficient may be calculated.

(Second Embodiment)
Next, the counterfeit-proof printed matter comprehensive evaluation system (B) according to the second embodiment of the present invention will be described. In the second embodiment, when there are mutual dependencies among a plurality of evaluation items set in the same hierarchy, for example, if the evaluation value of the evaluation item A is high, the evaluation of the evaluation item B is also high or Conversely, if there is a relationship such that if the evaluation value of the evaluation item A is low, the evaluation of the evaluation item B is also low, by correcting the weighting factor calculated by the pair comparison of the evaluation items, further accuracy can be obtained. This is a comprehensive evaluation system for anti-counterfeit printed matter (B) and a comprehensive evaluation method for anti-counterfeit printed matter that can be highly evaluated. A method of correcting the weighting factor and specific examples of the evaluation items that depend on each other will be described in the configuration of the forgery-preventing printed matter comprehensive evaluation system (B) and the counterfeit-preventing printed matter comprehensive evaluation method described below.

FIG. 13 is a block diagram showing the configuration of the forgery prevention print comprehensive evaluation system (B) according to the second embodiment of the present invention. As in the first embodiment, the counterfeit prevention printed matter comprehensive evaluation system (B) is roughly divided into a control unit (C), and is roughly divided into an evaluation tree structure setting unit (S), a data input unit (I), and an evaluation output unit. (O). The specific functions of each part will be described later, but the series of functions should be implemented using spreadsheet software or numerical analysis software that runs on a general-purpose personal computer in the same way as the comprehensive anti-counterfeit printed evaluation system (A). Can do.

The difference between the forgery-preventing printed matter comprehensive evaluation system (A) described in the first embodiment and the forgery-preventing printed matter comprehensive evaluation system (B) of the second embodiment is as shown in FIG. (I) includes a mutual dependency value input unit (I4), and the evaluation processing unit (P) includes a dependency coefficient calculation unit (P7) and a weight correction calculation unit (P8). In FIG. 13, these different parts are shown in different colors for easy understanding. Hereinafter, mainly different parts will be described and explained.

(Interdependent value input section)
The interdependent value input (I4) part is a feature of the second embodiment of the present invention, and is a part for inputting an evaluation result of interdependency between evaluation items. Although the details of the interdependency will be described later, when the input by the interdependence value input unit (I4) is briefly described, the degree of influence between the evaluation items belonging to the same hierarchy with respect to one evaluation item is evaluated. A pair comparison is performed to input the evaluation value. Note that the paired comparison value input in the interdependent value input unit (I4) may be stored in advance in the storage unit (M), read out, and input the corresponding value, or the paired comparison value may be input. At the stage of performing, a paired comparison value may be derived and input as appropriate.

(Dependency coefficient calculator)
The dependency coefficient calculation unit (P7) is a part that calculates the interdependence weight coefficient from the evaluation value by pair comparison between the respective evaluation items input by the interdependence value input unit (I4). In addition to the main eigenvector method described in (Equation 1), (Equation 2), (Equation 3), and (Equation 4), the calculation method of the interdependent weighting coefficient uses a geometric average method or a harmonic average method. In some cases, it is implemented on spreadsheet software or numerical analysis software.

(Weight correction calculation unit)
The weight correction calculation unit (P8) is calculated by the weighting coefficient of each evaluation item (having the mutual dependence of each evaluation item) derived in advance by the weighting coefficient calculation unit (P1) and the dependency coefficient calculation unit (P7). This is a part that performs multiplication with the interdependent weighting coefficient. The structure and function of the other parts are the same as those in the forgery prevention print comprehensive evaluation system (A).

Next, the comprehensive evaluation method for forgery prevention printed matter of the second embodiment using the forgery prevention printed matter comprehensive evaluation system (B) will be described. FIG. 14 is a flowchart illustrating an overview of the comprehensive evaluation method according to the second embodiment. The difference from the comprehensive evaluation method of the first embodiment is that a weighting factor correction step (G3) is newly added after the weighting factor calculation step (G2) in FIG. In order to facilitate understanding, the weighting coefficient correction step (G3) in FIG. 14 is shown in different colors. The correction of the weighting factor is a weighting factor calculation step (G2) for the purpose of eliminating a reduction in evaluation accuracy due to mutual dependency between evaluation items derived as a tree structure, as described in the problem in the prior art. The correction calculation is performed using the interdependent weighting coefficient calculated in the weighting coefficient correction step (G3) with respect to the weighting coefficient in each evaluation item calculated in (1). The detailed flow of the weight coefficient correction step (G3) in FIG. 14 is shown in FIG. 15 and will be described with reference to examples.

  The first step (G3-1) in FIG. 15 is an input of a pairwise comparison value of interdependency in each evaluation item. This is performed using the interdependent value input unit (I4). As a specific example, the evaluation time of the interdependency between each evaluation item will be described by exemplifying the identification time, the observation angle, and the identification illuminance, which are the three evaluation items in the discrimination cited in the first embodiment.

The outline of the interdependency evaluation is described. First, one evaluation item is taken out from each evaluation item, and a pairwise comparison is performed on the interdependency between the evaluation items in consideration of the one evaluation item. Next, the remaining evaluation items are also picked up sequentially, and finally, the same pairwise comparison between the evaluation items is performed for all the evaluation items. Here, a paired comparison method for interdependency between evaluation items in one evaluation item will be described.

For example, in the case of the identification time, observation angle, and identification illuminance, which are the three evaluation items relating to the above-described discrimination, first, a pair comparison between each evaluation item in the identification time, specifically, the identification time (pair) expression angle A pairwise comparison of the identification time (pair) expression illuminance and the expression angle (pair) identification illuminance is performed. Here, an example of the interdependency evaluation of the identification angle (vs.) identification illuminance in the identification time will be described. First, when performing a paired comparison, the paired comparison is performed from the viewpoint of which evaluation item has an influence on the identification time or the same degree with the identification angle and the identification illuminance. Here, the evaluation values shown in Table 20 are used as the paired comparison values.

(Table 20)

For example, in the technique described in Japanese Patent No. 2615401, an image cannot be visually recognized when viewed from the front, but a latent image appears by tilting. If the tilt angle at this time is small, the time required for tilting, That is, the identification time is short, and if the tilting angle is large, the time required for tilting, that is, the identification time is long. Therefore, it can be said that the identification time depends on the magnitude of the expression angle. Similarly, the identification time depends on the observation environment of the technique described in Japanese Patent No. 26154011, for example, the identification illuminance. In an observation environment that does not have sufficient brightness from the viewpoint of recognition in ergonomics, that is, an environment below 400 LX, more identification time is required. In other words, the interdependent pair comparison is to compare the intensity of the influence of the observation angle and the identification illuminance during the identification time.

Table 21 shows the results of a pairwise comparison of the influence of each evaluation item, that is, the evaluation of the strength of interdependence between the evaluation items. As described above, one evaluation item is sequentially picked up from the three evaluation items regarding discriminability, and a pairwise comparison of the interdependencies between the evaluation items is performed in view of the picked evaluation items. Therefore, Table 21 shows three pairs of comparison results. The result of the pair comparison was input using the interdependent value input unit (I4) in FIG. The method for inputting the paired comparison value is based on paragraph (0010). If the pairing comparison of the influence on the identification time shown in Table 21 is performed, a pairwise comparison of the expression angle and the identification illuminance is performed. If the expression angle is evaluated as having a stronger influence than the discriminating illuminance (from “Strong... 5” Table 20), 5 is inserted in the row direction of the expression angle. Along with this, 1/5, which is the reciprocal of the numerical value of the previous operation, is automatically inserted in the row direction of the identification illuminance. It should be noted that 1 is inserted on the diagonal line of the pair comparison result shown in Table 21 because it is the comparison result of the same evaluation item.

(Table 21)

The second step (G3-2) in FIG. 15 is the calculation of the interdependent weight coefficient. The interdependence weight coefficient indicates the strength of the interdependence between the evaluation items derived from the paired comparison result between the evaluation items in the specific evaluation item described in the previous section. This calculation method is based on the main eigenvector method described in (Equation 1), (Equation 2), (Equation 3), and (Equation 4), using the dependency coefficient calculation unit (P7) in the control unit (C). Calculated. It should be noted that the calculation of the interdependent weighting factor can be assumed to use other methods using a geometric average or a harmonic average. As a specific example, Table 22 shows the calculation results of the interdependent weighting factors in the three evaluation items shown in Table 21 inputted in the first step (G3-1). Note that the interdependent weighting coefficient is calculated in the form of 3 rows and 1 column for each evaluation item, but in the figure, the interdependent weighting factors are collected and displayed in 3 columns in the horizontal direction.

(Table 22)

  The third step (G3-3) in FIG. 15 is a calculation of the degree of consistency of the interdependent weighting coefficient calculated by the dependency coefficient calculating unit (P7). The method for calculating the degree of matching is calculated by the degree-of-matching calculation determination unit (P2) in the control unit (C) of FIG. 13 based on the equation described in (Expression 5). As a specific example, the matching degrees of the interdependent weight coefficients shown in Table 22 were calculated as 0.006, 0.019, and 0.000, respectively, in the same order of the identification time, the observation angle, and the identification illuminance.

  The fourth step (G3-4) in FIG. 15 is determination of the calculated degree of matching. If it exceeds 0.15, since there is a concern that the weighting includes blurring and fluctuation, the process returns to the first step (G3-1) as shown by the arrow in FIG. Re-enter paired comparison values for interdependencies. The three values obtained in this specific example are all less than 0.15, and therefore all determined to be appropriate. In addition, as shown in FIG. 15, the calculated | required interdependence weighting coefficient can be suitably preserve | saved in the memory | storage part (M) in a control part (C) as needed.

  The fifth step (G3-5) in FIG. 15 is reading of the weighting factor stored in the storage unit (M) in the control unit (C). This weighting factor is calculated based on a paired comparison regarding the importance (or priority) of each evaluation item in the weighting factor calculation step (G2) of FIG. 14, and is stored in advance. As an example in the present embodiment, the weighting coefficient shown in Table 10 is used.

  The sixth step (G3-6) in FIG. 15 is the multiplication of the interdependent weighting factor. This is obtained by multiplying the weighting factor read in the fifth step (G3-5) by the interdependence weighting factor calculated in the second step (G3-2), and thereby interdependence between the evaluation items. It is intended to eliminate the nature. The multiplication of the interdependent weight coefficient is performed using the weight correction calculation unit (P8) of the control unit (C). When the interdependent weighting factor is mi (i = 1, 2, 3,..., N) and the previously determined weighting factor is wi (i = 1, 2, 3,..., N), The multiplication result Wi (i = 1, 2, 3,..., N) of the dependency weight coefficient is expressed by the following equation (10). However, n represents the number of evaluation items.

(Equation 10)

  An example of the correction calculation using the weight correction calculation unit (P8) in the control unit (C) based on the multiplication defined in (Formula 10) is shown in (Formula 11). The weighting factor wi used here is that shown in Table 10 of the first embodiment, and the interdependence factor mi is that shown in Table 22. The weight coefficient Wi after the correction calculation was calculated as follows.

・・・（１１）

(Equation 11)

(11)

  Note that the values (0.264, 0.459, 0.277) of W1, W2, and W3 corrected by (Equation 11) are as shown in the seventh step (G3-7) in FIG. And stored in the storage unit (M) in the control unit (C).

  The weight coefficient correction step (G3) in the comprehensive evaluation method for forgery-preventing printed matter according to the second embodiment has been described above in detail. As described above, the first (G1), second (G2), fourth (G4), fifth (G5), sixth (G6) and seventh (G7) steps in FIG. However, in the partial performance evaluation value calculation step (G5), the partial performance evaluation raw value is multiplied by the corrected weighting coefficient. FIG. 16 shows an evaluation result of the comprehensive performance evaluation value and the partial performance evaluation value relating to the discriminability newly calculated using the corrected weighting coefficient of each evaluation item as an example.

  In the second embodiment, the evaluation value shown in FIG. 16 is obtained with respect to the evaluation result shown in FIG. 12 of the first embodiment by using a weighting factor in consideration of the interdependency between the evaluation items. Is different. From this, in the second embodiment, when there is a dependency relationship between the evaluation items, the corrected weighting factor is reflected, and it can be seen that appropriate evaluation can be performed.

A Comprehensive evaluation system for anti-counterfeit printed matter S Evaluation tree structure setting unit S1 Hierarchy setting unit S2 Evaluation item input unit I Data input unit I1 Weight input unit I2 Performance level value input unit I3 Evaluation value input unit I4 Interdependent value input unit C Control Part P Evaluation processing part P1 Weight coefficient calculation part P2 Consistency calculation calculation determination part P3 Performance weight coefficient set calculation part P4 Partial performance evaluation elementary value calculation part P5 Partial performance evaluation value calculation part P6 Total performance evaluation value calculation part P7 Dependency coefficient calculation part P8 Weight correction calculation unit M Storage unit O Evaluation value output unit O1 Evaluation list creation unit O2 Evaluation profile creation unit

Claims (10)

A method for evaluating the overall performance of an anti-counterfeit printed matter using a numerical value measured from the anti-counterfeit printed matter with respect to the evaluation items as a plurality of requirements in the anti-counterfeit printed matter. And a pair of evaluation items in the same hierarchy set in the tree structure setting step and a tree structure setting step for setting, as evaluation items, a plurality of requirements for realizing a higher-level goal. A weighting factor calculating step for calculating a weighting factor of the evaluation item from the first eigenvector by comparison, and for each of the set evaluation items, a numerical value measured from the anti-counterfeit printed matter to be evaluated is classified A plurality of levels are set, and a performance weight coefficient for each level is obtained from a second eigenvector obtained by pairwise comparison between the levels set in each evaluation item. In the performance weight coefficient set calculation step, the performance weight coefficient set calculation step calculates and sets the numerical value measured from the forgery prevention printed matter to be evaluated for each of the set evaluation items. Classify into one of a plurality of set levels, and divide the performance weight coefficient corresponding to the classified level by the maximum value of the performance weight coefficient constituting the same group to obtain a partial performance evaluation element A partial performance evaluation raw value calculation step for calculating a value, the partial performance evaluation raw value calculated in the partial performance evaluation raw value calculation step, and the weight coefficient corresponding to the partial performance evaluation raw value A partial performance evaluation value calculating step for calculating a performance evaluation value; and a total performance evaluation value calculating step for performing a total calculation of the partial performance evaluation values calculated in the partial performance evaluation value calculation step. How to evaluate the overall performance of the anti-counterfeit printed matter characterized by. In the evaluation items in the same hierarchy set in the tree structure setting step, an interdependent weight coefficient is calculated from a third eigenvector by pair comparison for evaluating the degree of influence of each evaluation item on one evaluation item, 2. The forgery-preventing printed matter according to claim 1, further comprising a weighting factor correcting step for multiplying the weighting factor by the interdependent weighting factor between the weighting factor calculating step and the partial performance evaluation value calculating step. To evaluate the overall performance of the system. A system for evaluating the overall performance of an anti-counterfeit printed matter by using a plurality of requirements in the anti-counterfeit printed matter as evaluation items, and using numerical values measured from the anti-counterfeit printed matter for the evaluation items, A tree structure setting unit for setting a hierarchical structure with a plurality of requirements as evaluation items, and a numerical value measured from an anti-counterfeit printed matter to be evaluated in each evaluation item set in the tree structure setting unit, the measurement At least a paired comparison value used for each paired comparison is input at a plurality of levels for classifying the numerical values to be classified, between the levels and between evaluation items in the same hierarchy set by the tree structure setting unit A weighting coefficient is calculated from a data input unit and eigenvectors of each paired comparison value input at the input unit, and each evaluation item is calculated. Classifying numerical values measured from the anti-counterfeit printed matter to be evaluated, which are input at the input unit, into one of the plurality of levels, and the same weighting factor corresponding to the classified level Dividing by the maximum value of the weighting coefficient at the level of the plurality of evaluation items, multiplying the divided value and the weighting coefficient calculated in each evaluation item, respectively, to calculate the evaluation value of each evaluation item, A counterfeit-prevention printed matter comprehensive performance evaluation system, comprising: a control unit including an evaluation processing unit that performs an overall evaluation by adding the evaluation values of the respective evaluation items. The tree structure setting unit has a hierarchical setting unit that sets at least two or more hierarchical structures for setting a plurality of the evaluation items for realizing a higher-level objective as a final purpose. 4. A forgery-preventing printed matter comprehensive performance evaluation system according to claim 3, further comprising an evaluation item input unit for setting the evaluation item in the set hierarchical structure. The data input unit includes a performance level input unit that inputs values of intervals of the plurality of levels, a weight input unit that inputs paired comparison values used for pairwise comparison between the levels and the evaluation items, and the evaluation 5. The total performance evaluation system for anti-counterfeit printed matter according to claim 3 or 4, further comprising an evaluation value input unit for inputting a numerical value measured from the anti-counterfeit printed matter that is the object of the forgery. The evaluation processing unit is a weighting factor calculation unit that calculates a weighting factor for each evaluation item, a performance weighting factor set calculation unit that calculates the weighting factors of the plurality of levels for each of the evaluation items, and sets them as one group In each of the evaluation items, the numerical value measured from the anti-counterfeit printed matter to be evaluated is classified into one of the plurality of levels, and the maximum value of the weighting factor that constitutes the same one group. Partial performance evaluation elementary value calculation unit for dividing, evaluation of each evaluation item calculated by the partial performance evaluation value calculation unit for multiplying the divided value and the weight coefficient of each evaluation item, and the partial performance evaluation value calculation unit, respectively 6. The system for evaluating overall performance of forgery-preventing printed matter according to claim 3, further comprising an overall performance evaluation value calculation unit for summing values. The data input unit is a part to which a pair comparison value used for pair comparison for evaluating the degree of influence of each evaluation item with respect to one evaluation item is input in the evaluation item in the same hierarchy, and the evaluation The processing unit calculates a weighting factor from an eigenvector by a paired comparison value that evaluates the degree of influence of each evaluation item on one evaluation item in the evaluation items in the same hierarchy input by the data input unit, and The total performance evaluation system for anti-counterfeit printed matter according to any one of claims 3 to 6, wherein the weighting factor calculated in the item is multiplied for correction. The data input unit has an interdependent value input unit for inputting a paired comparison value used for paired comparison for evaluating the degree of influence of each evaluation item on one evaluation item in the evaluation items in the same hierarchy. The evaluation processing unit includes a dependency coefficient calculation unit that calculates a weighting factor from an eigenvector based on a pair of comparison values input by the interdependent value input unit, and the dependency coefficient to the weighting factor calculated in each evaluation item. The total performance evaluation system for forgery-preventing printed matter according to claim 7, further comprising a weight correction calculation unit that multiplies and corrects the weighting factor calculated by the calculation unit. An evaluation output unit that creates and outputs in a list and / or a predetermined format a total evaluation value obtained by adding together the evaluation value of each evaluation item and the evaluation value of each evaluation item in the anti-counterfeit printed matter to be evaluated The overall performance evaluation system for anti-counterfeit printed matter according to claim 3, further comprising: The evaluation output unit is configured to generate a list of evaluation values of the respective evaluation items and a value of the comprehensive evaluation, and to output the evaluation list generation unit. The evaluation value of the evaluation items and the value of the comprehensive evaluation are predetermined. The total performance evaluation system for anti-counterfeit printed matter according to claim 9, further comprising an evaluation profile creation unit that creates and outputs in a format.

Images