Method of and system for analyzing, modeling and valuing elements of a business enterprise

• US 20080071588A1
• Filed: 08/27/2001
• Published: 03/20/2008
• Est. Priority Date: 12/10/1997
• Status: Abandoned Application

• Current Assignee: Jeffrey Eder
• Original Assignee: Jeffrey Eder
• Inventors: Eder, Jeff S.
• Application Number:
  US09/938,874
• Publication Number:
  US 20080071588A1
• Time in Patent Office: Days
• US Class Current: 705/7
• CPC Class Codes: G06Q 10/06 Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
• CPC Class Codes: G06Q 10/063 Operations research, analysis or management
• CPC Class Codes: G06Q 10/06375 Prediction of business process outcome or impact based on a proposed change
• CPC Class Codes: G06Q 30/02 Marketing; Price estimation or determination; Fundraising
• CPC Class Codes: G06Q 30/0202 Market predictions or forecasting for commercial activities
• CPC Class Codes: G06Q 40/00 Finance; Insurance; Tax strategies; Processing of corporate or income taxes
• CPC Class Codes: G06Q 40/02 Banking, e.g. interest calculation or account maintenance credit or loans G06Q40/03
• CPC Class Codes: G06Q 40/12 Accounting

Abstract

An automated system (100) and method for analyzing, modeling and valuing elements of a business enterprise on a specified valuation date. The performance of the elements are analyzed using search algorithms and induction algorithms to determine the value drivers associated with each element. The induction algorithms are also used to create composite variables that relate element performance to enterprise revenue, expenses and changes in capital. Predictive models are then used to determine the correlation between the value drivers and the enterprise revenue, expenses and changes in capital. The correlation percentages for each value driver are then multiplied by capitalized value of future revenue, expenses and changes in capital, the resulting numbers for each value driver associated with each element are then added together to calculate a value for each element.

Claims

1-78. -78. (canceled)

79. A computer based method of building predictive models from transaction data, comprising:

aggregating data from a plurality of transaction systems covering a series of time periods for one or more elements of value and one or more aspects of financial performance;

transforming said element of value data in accordance with one or more pre-programmed functions;

establishing a plurality of input nodes, a plurality of hidden nodes and an output node for a neural network model for each aspect of financial performance;

inputting the raw and transformed transaction data into each neural network model using a separate input node for untransformed transaction data and each pre-programmed transformation function by element of value for all time periods in the series;

training each neural network model using said inputs until an error function associated with an output value that corresponds to an aspect of financial performance is minimized; and

using one or more weights from the trained neural network models to identify a set of raw and transformed transaction data by element of value and output that will be used as an input to an element of value summary for each of one or more predictive models

normalizing each of the one or more sets of raw and transformed transaction data by element of value,

refining the sets of raw and transformed transaction data by element of value, creating a summary of the refined transaction data set for each element of value, and using the element of value summaries as inputs to a predictive model for each of the one or more aspects of enterprise financial performance

where the aspects of financial performance are selected from the group consisting of revenue, expense, capital change, cash flow and combinations thereof, and

where the predictive models of aspects of financial performance are useful for completing tasks selected from the group consisting of optimizing a current operation financial performance for a business, predicting an impact of one or more changes to a current operation financial performance, calculating a value for an element of value and combinations thereof.

80. The method of claim 79 where a plurality of input nodes is set equal to one plus the number of elements of value times one plus the number of pre-programmed functions used to transform transaction data.

81. The method of claim 79 where a plurality of hidden nodes is set equal to one plus the number of input nodes.

82. The method of claim 79, where an error function further comprises ERR(W)k=½

(Rk−

Y(W))².

83. The method of claim 79 where a set of raw and transformed transaction data that will be used as an input to a predictive model further comprises a set of numbers.

84. The method of claim 79 where one or more predictive models further comprises one or more neural network models.

85. The method of claim 79 where training a neural network model further comprises using a genetic algorithm to complete the training where a population being analyzed is partitioned into a plurality of subpopulations, with each subpopulation being processed by a genetic algorithm independently of the others and where a migration mechanism produces a chromosome exchange between the subpopulations.

86. The method of claim 79 where training a neural network model further comprises using a back propagation algorithm to complete the training.

87. The method of claim 79 where one or more elements of value further comprise one or more elements of value selected from the group consisting of brands, customers, employees, partners, vendors and combinations thereof.

88. The method of claim 79 where a plurality of transaction systems comprise systems selected from the group consisting of advanced financial systems, basic financial systems, operation management systems, sales management systems, human resource systems, accounts receivable systems, accounts payable systems, capital asset systems, inventory systems, invoicing systems, payroll systems, purchasing systems, the Internet and combinations thereof.

89. The method of claim 79 where a series of time periods contains time periods selected from the group consisting of historical time periods, future time periods and combinations thereof.

90. The method of claim 79 where the one or more pre programmed functions are selected from the group consisting of average, rolling average, time delay, trend, average time delay, rolling average time delay, ratio, average ratio, rolling average ratio, slope, average slope, rolling average slope and combinations thereof.

91. The method of claim 79 where training a neural network model further comprises using a genetic algorithm to complete the training where a population being analyzed is partitioned into a plurality of subpopulations, with each subpopulation being processed by a genetic algorithm independently of the others and where a selective crossover and a fitness measure rescaling produces a chromosome exchange between the subpopulations.

92. A program storage device readable by a computer, tangibly embodying a program of instructions executable by at least one computer to perform the steps in method, comprising:

aggregating data from a plurality of transaction systems covering a series of time periods for one or more elements of value and one or more aspects of financial performance;

transforming said element of value data in accordance with one or more pre-programmed functions;

establishing a plurality of input nodes, a plurality of hidden nodes and an output node for a neural network model for each aspect of financial performance;

inputting the raw and transformed transaction data into each neural network model using a separate input node for untransformed transaction data and each pre-programmed transformation function by element of value for all time periods in the series;

training each neural network model using said inputs until an error function associated with an output value that corresponds to an aspect of financial performance is minimized; and

using one or more weights from the trained neural network models to identify a set of raw and transformed transaction data by element of value and output that will be used as an element of value summary for use as an input to each of one or more predictive models

normalizing each of the one or more sets of raw and transformed transaction data by element of value, refining the sets of raw and transformed transaction data by element of value,

creating a summary of the refined transaction data set for each element of value, and

using the element of value summaries as inputs to a predictive model for each of the one or more aspects of enterprise financial performance

where the aspects of financial performance are selected from the group consisting of revenue, expense, capital change, cash flow and combinations thereof, and

where the predictive models of aspects of financial performance are useful for completing tasks selected from the group consisting of optimizing a current operation financial performance for a business, predicting an impact of one or more changes to a current operation financial performance, calculating a value for an element of value and combinations thereof.

93. The program storage device computer readable medium of claim 92 where a plurality of input nodes is set equal to one plus the number of elements of value times one plus the number of pre-programmed functions used to transform transaction data.

94. The program storage device of claim 92 where a plurality of hidden nodes is set equal to one plus the number of input nodes.

95. The program storage device of claim 92, where an error function further comprises ERR(W)k=½

(Rk−

Y(W))².

96. The program storage device of claim 92 where a set of raw and transformed transaction data that will be used as an input to a predictive model further comprises a set of numbers.

97. The program storage device of claim 92 where one or more predictive models further comprises one or more neural network models.

98. The program storage device of claim 92 where training a neural network model further comprises using a genetic algorithm to complete the training where a population being analyzed is partitioned into a plurality of subpopulations, with each subpopulation being processed by a genetic algorithm independently of the others and where a migration mechanism produces a chromosome exchange between the subpopulations.

99. The program storage device of claim 92 where training a neural network model further comprises using a back propagation algorithm to complete the training.

100. The program storage device computer readable medium of claim 92 where one or more elements further comprise elements of value selected from the group consisting of brands, customers, employees, partners, vendors and combinations thereof.

101. The program storage device of claim 92 where a plurality of transaction systems comprise systems selected from the group consisting of advanced financial systems, basic financial systems, operation management systems, sales management systems, human resource systems, accounts receivable systems, accounts payable systems, capital asset systems, inventory systems, invoicing systems, payroll systems, purchasing systems, the Internet and combinations thereof.

102. The program storage device of claim 92 where a series of time periods contains time periods selected from the group consisting of historical time periods, future time periods and combinations thereof.

103. The program storage device of claim 92 where the one or more pre programmed functions are selected from the group consisting of average, rolling average, time delay, trend, average time delay, rolling average time delay, ratio, average ratio, rolling average ratio, slope, average slope, rolling average slope and combinations thereof.

104. The program storage device of claim 92 where the method further comprises:

training a neural network by using a genetic algorithm to complete the training where a population being analyzed is partitioned into a plurality of subpopulations, with each subpopulation being processed by a genetic algorithm independently of the others and where a selective crossover and a fitness measure rescaling produces a chromosome exchange between the subpopulations.

105. An apparatus for building predictive models from transaction data, comprising:

a plurality of transaction systems,

means for preparing data from said systems for use in processing for a series of time periods for one or more elements of value and one or more aspects of financial performance;

means for transforming said element of value data in accordance with one or more pre-programmed functions;

means for establishing a plurality of input nodes, a plurality of hidden nodes and an output node for a neural network model for each aspect of financial performance;

means for inputting the raw and transformed transaction data into each neural network model using a separate input node for untransformed transaction data and each pre-programmed transformation function by element of value for all time periods in the series;

means for training each neural network model using said inputs until an error function associated with an output value that corresponds to an aspect of financial performance is minimized; and

means for using one or more weights from the trained neural network models to identify a set of raw and transformed transaction data by element of value and output that will be used as an element of value summary for use as an input to each of one or more predictive models

normalizing each of the one or more sets of raw and transformed transaction data by element of value,

refining the sets of raw and transformed transaction data by element of value, creating a summary of the refined transaction data set for each element of value, and

using the element of value summaries as inputs to a predictive model for each of the one or more aspects of enterprise financial performance

where the aspects of financial performance are selected from the group consisting of revenue, expense, capital change, cash flow and combinations thereof.

106. The apparatus of claim 105 where a plurality of input nodes is set equal to one plus the number of elements times one plus the number of pre-programmed functions used to transform transaction data.

107. The apparatus of claim 105 where a plurality of hidden nodes is set equal to one plus the number of input nodes.

108. The apparatus of claim 105, where an error function further comprises ERR(W)k=½

(Rk−

Y(W))².

109. The apparatus of claim 105 where a set of raw and transformed transaction data that will be used as an input to a predictive model further comprises a set of numbers.

110. The apparatus of claim 105 where one or more predictive models further comprises one or more neural network models.

111. The apparatus of claim 105 where training a neural network model further comprises using a genetic algorithm to complete the training where a population being analyzed is partitioned into a plurality of subpopulations, with each subpopulation being processed by a genetic algorithm independently of the others and where a migration mechanism produces a chromosome exchange between the subpopulations.

112. The apparatus of claim 105 where training a neural network model further comprises using a back propagation algorithm to complete the training.

113. The apparatus of claim 105 where one or more elements further comprise elements of value selected from the group consisting of brands, customers, employees, partners, vendors and combinations thereof.

114. The apparatus of claim 105 where one or more outputs further comprise a plurality of transaction systems comprise systems selected from the group consisting of advanced financial systems, basic financial systems, operation management systems, sales management systems, human resource systems, accounts receivable systems, accounts payable systems, capital asset systems, inventory systems, invoicing systems, payroll systems, purchasing systems, the Internet and combinations thereof.

115. The apparatus of claim 105 where a series of time periods contains time periods selected from the group consisting of historical time periods, future time periods and combinations thereof.

116. The apparatus of claim 105 where the one or more pre programmed functions are selected from the group consisting of average, rolling average, time delay, trend, average time delay, rolling average time delay, ratio, average ratio, rolling average ratio, slope, average slope, rolling average slope and combinations thereof.

117. The apparatus of claim 105 where preparing data for use in processing further comprises integrating, converting and storing data from a plurality of systems in accordance with a common data dictionary.

118. The apparatus of claim 105 that further comprises:

means for training a neural network by using a genetic algorithm to complete the training where a population being analyzed is partitioned into a plurality of subpopulations, with each subpopulation being processed by a genetic algorithm independently of the others and where a selective crossover and a fitness measure rescaling produces a chromosome exchange between the subpopulations.

119. A data processing method, comprising:

organizing business transaction data by enterprise into one or more components of value and two or more elements of value where at least one element of value is intangible;

determining a relative contribution of each of two or more elements of value to a value of a business by analyzing at least a portion of the data; and

reporting the relative contribution of each element of value and the value of the business.

120. The data processing method of claim 119 wherein determining a relative contribution for each of the two or more elements to a value of the business further comprises:

deriving an element of value weighting factor for each element of value by enterprise; and

weighting the data concerning each of one or more elements of value according to the element of value weighting factors for each enterprise, where the relative value contribution is the sum of the weighted element of value data for all enterprises within the business.

121. The data processing method of claim 119 wherein the intangible elements of value are selected from the group consisting of brands, customers, employees, partners, vendors and combinations thereof.

122. The data processing method of claim 119 wherein reporting the value of the business and the relative contribution by element of value further comprises the use of a paper document or an electronic display.

123. The data processing method of claim 119 wherein a value of the business is market value.

124. The data processing method of claim 119 wherein at least one of the two or more elements of value contain items that are optionally clustered into sub-elements of value for more detailed analysis.

125. The data processing method of claim 119 wherein transaction data is obtained from the group consisting of advanced financial systems, basic financial systems, operation management systems, sales management systems, human resource systems, accounts receivable systems, accounts payable systems, capital asset systems, inventory systems, invoicing systems, payroll systems, purchasing systems and combinations thereof.

126. The data processing method of claim 119 wherein at least a portion of the data is obtained from the Internet.

127. The data processing method of claim 119 wherein an enterprise is defined by a revenue component of value together with an optional component of value selected from the group consisting of expense, capital change and combinations thereof.

128. The data processing method of claim 119 wherein a revenue component of value that defines an enterprise can include the revenue from a single product, a group of products, a division or an entire company.

129. The data processing method of claim 119 wherein each of one or more components of value can be divided into subcomponents of value for more detailed analysis.

130. The data processing method of claim 119 wherein a relative contribution for each element of value further comprises a relative contribution for a specified point in time within a sequential series of points in time.

131. The data processing method of claim 119 wherein a relative contribution of each element of value to a value of a business is determined by a relative impact of the element of value on the components of value and the other elements of value by enterprise.

132. The data processing method of claim 120 wherein deriving one or more element of value weighting factors further comprises:

determining an initial weighting factor with a predictive neural network model; and

finalizing the element of value weighting factors with a model selected from the group consisting of entropy minimization, lagrange and path analysis.

133. The data processing method of claim 132 wherein the method further comprises using genetic algorithms to evolve each model to an optimal configuration before completing each method step.

134. A program storage device having sequences of instructions stored therein, which when executed causes the processor in a computer to perform a data processing method, comprising:

organizing business data by enterprise into one or more components of value and two or more elements of value where at least one element of value is intangible;

determining a relative contribution of each of two or more elements of value to a value of the business by analyzing at least a portion of the data; and

reporting the relative contribution of each element of value and the value of the business.

135. The program storage device of claim 134 wherein determining a relative contribution for each of the two or more elements to a value of the business further comprises:

deriving an element of value weighting factor for each element of value by enterprise; and

weighting the data concerning each of one or more elements of value according to the element of value weighting factors for each enterprise, where the relative value contribution is the sum of the weighted element of value data for all enterprises within the business.

136. The program storage device of claim 134 wherein an intangible element of value is selected from the group consisting of brands, customers, employees, partners, vendors and combinations thereof.

137. The program storage device of claim 134 wherein reporting the value of the business and the relative contribution by element of value further comprises the use of a paper document or an electronic display.

138. The program storage device of claim 134 wherein a value of the business is market value.

139. The program storage device of claim 134 wherein at least one of the two or more elements of value contain items that are optionally clustered into sub-elements of value for more detailed analysis.

140. The program storage device of claim 134 wherein transaction data is obtained from the group consisting of advanced financial systems, basic financial systems, operation management systems, sales management systems, human resource systems, accounts receivable systems, accounts payable systems, capital asset systems, inventory systems, invoicing systems, payroll systems, purchasing systems and combinations thereof.

141. The program storage device of claim 134 wherein at least a portion of the data is obtained from the Internet.

142. The program storage device of claim 134 wherein an enterprise is defined by a revenue component of value together with an optional component of value selected from the group consisting of expense, capital change and combinations thereof.

143. The program storage device of claim 134 wherein a revenue component of value that defines an enterprise can include the revenue from a single product, a group of products, a division or an entire company.

144. The program storage device of claim 134 wherein each of one or more components of value can be divided into subcomponents of value for more detailed analysis.

145. The program storage device of claim 134 wherein a relative contribution for each element of value further comprises a relative contribution for a specified point in time within a sequential series of points in time.

146. The program storage device of claim 134 wherein a relative contribution of each element of value to a value of a business is determined by a relative impact of the element of value on the components of value and the other elements of value by enterprise.

147. The program storage device of claim 135 wherein deriving one or more element of value weighting factors further comprises:

determining an initial weighting factor with a predictive neural network model; and

finalizing the element of value weighting factors with a model selected from the group consisting of entropy minimization, lagrange and path analysis.

148. The program storage device of claim 147 wherein the method further comprises using genetic algorithms to evolve each model to an optimal configuration before completing each method step.

149. A financial system, comprising:

networked computers each with a processor having circuitry to execute instructions;

a storage device available to each processor with sequences of instructions stored therein, which when executed cause the processors to;

integrate transaction data from a plurality of enterprise management systems, analyze at least a portion of the integrated data to identify one or more events that drive enterprise value creation and a business context that is associated with said events, and

using transaction data associated with said events to develop a computational model of enterprise financial performance.

150. The system of claim 149 wherein a computational model further comprises up to three network component of value models where a plurality of tangible and intangible elements of value are connected to a level of each component of value over time and where automated analysis through computational techniques is supported.

151. The system of claim 149 wherein a computational model further comprises a causal model that supports automated analysis through computational techniques.

152. The system of claim 149 wherein one or more intangible elements of value are selected from the group consisting of brands, customers, employees, intellectual capital, partners, vendors, vendor relationships and combinations thereof.

153. The system of claim 149 wherein one or more tangible elements of value further comprise production equipment.

154. The system of claim 149 where one or more components of value are selected from the group consisting of revenue, expense, capital change and combinations thereof.

155. The system of claim 149 that produces useful results selected from the group consisting a valuation for one or more elements of value, an impact quantification for one or more elements of value, one or more financial performance forecasts that do not require reconciliations, one or more forecasts of the expected impact of change to an element value driver and combinations thereof.

156. A program storage device readable by a computer, tangibly embodying a program of instructions executable by at least one computer to perform the steps in a neural network development method, comprising:

a) preparing a plurality of input data and ouput data for a population for use in neural network processing,

b) defining a structure for a neural network comprising a plurality of input nodes, a plurality of hidden nodes, an output node, a connection between each input node and each hidden node and a connection between each hidden node and the output node,

c) assigning a random weight value to the connections between each node and a target fitness level

d) creating a plurality of chromosomes that encode the weights between each node,

e) generating a successor set of weight values from said initial set of weight values by evolving the chromosomes with a genetic algorithm, the input data and the output data until the target fitness level is achieved,

f) implementing said neural network with the set of weight values that achieved the target fitness level

where the population being analyzed is partitioned into a plurality of subpopulations,

with each subpopulation being processed by a genetic algorithm independently of the others and where a selective crossover produces a chromosome exchange between the subpopulations, and

where the selective crossover occurs between two or more successive generations.

157. The program storage device of claim 156, wherein a neural network model connects one or more elements of value of a business enterprise to one or more aspects of financial performance of said business enterprise,

where each input node represents an element of value,

where each output node represents an aspect of financial performance,

where the weights between nodes represent a plurality of relationships where each relationship is a function of the impact of an element of value on other elements of value or on an aspect of financial performance, and

where one or more aspects of financial performance are selected from the group consisting of revenue, expense, capital change, market value and combinations thereof.

158. The program storage device of claim 156, wherein a neural network model further comprises a business event network model.

159. A computer implemented neural network modeling method, comprising:

a) preparing a plurality of input data and ouput data for a population for use in neural network processing,

b) defining a structure for a neural network comprising a plurality of input nodes, a plurality of hidden nodes, an output node, a connection between each input node and each hidden node and a connection between each hidden node and the output node,

c) assigning a random weight value to the connections between each node and a target fitness level

d) creating a plurality of chromosomes that encode the weights between each node,

e) generating a successor set of weight values from said initial set of weight values by evolving the chromosomes with a genetic algorithm, the input data and the output data until the target fitness level is achieved,

f) implementing said neural network with the set of weight values that achieved the target fitness level

where the population being analyzed is partitioned into a plurality of subpopulations,

with each subpopulation being processed by a genetic algorithm independently of the others and where a selective crossover produces a chromosome exchange between the subpopulations, and

where the selective crossover occurs between two or more successive generations.

160. The method of claim 159, wherein a neural network model connects one or more elements of value of a business enterprise to one or more aspects of financial performance of said business enterprise,

where each input node represents an element of value,

where each output node represents an aspect of financial performance,

where the weights between nodes represent a plurality of relationships where each relationship is a function of the impact of an element of value on other elements of value or on an aspect of financial performance, and

where one or more aspects of financial performance are selected from the group consisting of revenue, expense, capital change, market value and combinations thereof.

161. The method of claim 159, wherein a neural network model further comprises a business event network model.