These key value drivers are The expressway networks play an important role in Thailand varied for different industries or enterprises. Identifying key economic development, especially in enhancing the logistics value drivers can be derived from efforts to model the systems.
The Expressway Authority of Thailand, a state determinants key value drivers of firm profit within an enterprise, is responsible for project planning, managing the industry. Value creation is the process of enterprise risk construction, operating, and maintaining expressway networks. Due to high-stakes relationship, which can visualize scenarios via the causal model investment caused by a large number of uncertainty factors or or cause and effect diagram.
Due to the problems of identifying the risks during the operation phase, the enterprise risk relationships of the risk factors and the subjectivity of risk management, performance measurement, and future plans management and ineffectiveness of some methodologies, such should be well-prepared.
Otherwise, value is destroyed i. A company during operation. Modeling scenarios via Bayesian networks is the can increase its economic profit in the following ways [1]: method for uncertainty reasoning and knowledge representation that was advanced at the end of the 20th century. After establishing all the variables in a model, one must deliberately associate the variables that cause changes in 3. In Key value drivers or key success factors can be regarded as key general, a Bayesian network describes the joint probability risk factors if they cause unexpected value and loss opportunity distribution for a set of variables.
The network or graph loss and damage. Value drivers are varied for different visualization represents the cause-and-effect relations among industries or enterprises. They are variables influenced by variables, pointed out by arcs. The degree of relationship is management strategies and decisions that can significantly affect interpreted in terms of conditional probabilities according to the competitive advantage of the various firms in an industry.
Bayes theorem. Bayesian Networks allow stating conditional More specifically, identifying key value drivers can be derived independence assumptions that apply to subsets of the variables, from efforts to model the determinants of firm profit within an providing more tractable and less constraining than the global industry.
There are a number of strategic models that can be assumption of conditional independence. The output of Bayesian used to illustrate the role of value drivers.
The value of development and diversification. The concept of value drivers enterprise is the economic efficiency of enterprise and economic also suggests causal relationships between resources and performance on the basis of their overall operations of the organizational value creation.
Traditionally, those resources overall judgment. In general, enterprise core performance is the were physical or financial capital. Nowadays, the concept of financial benefits. And the value of enterprise is based on intellectual capital has been identified as a key resource and current and the past profits ability of enterprise and development driver of organizational performance and value creation.
And potential [3]. The most common method for measuring how resources interact to create value or how intangible assets shareholder value is Economic Value Added EVA or are converted into tangible outcomes can be visualized via Economic Profit EP.
This paper thus applies the concept of strategy map creation from the economic points of view. The most attractive for assessing key value drivers or key risk factors from both the advantage of EP compared with traditional profit measures is financial and nonfinancial dimensions. The opportunity costs explain why shareholders invest in a certain company instead of the other company. A where EP equals to net operating profits minus a charge for cost Bayesian Network is represented by a directed acyclic graph DAG , associated with sets of local conditional probabilities of invested capital, attached to each node, called Conditional Probability Table or NOPAT denotes net operating profit after taxes, CPT [4].
The network arcs represent the assertion that the Capital denotes total capital of a company, variable labeled in each node is conditionally independent of its nondescendants in the network given its immediate predecessors Adj Nopat and Adj Capital denote the adjustments made by in the network.
The methods to construct Bayesian networks can be majorly classified into two categories: i top-down modeling WACC refers to weighted average cost of capital, methods, and ii reverse-engineering methods. Top-down including capital invested by shareholders and creditors.
There are four types of posterior approaches utilize machine learning algorithms to train learn probability inference defined as following [6]: Bayesian network structure and parameters from a collection of 1 Diagnostic inference, which can infer the causes in the light past observations. This process belongs to unsupervised learning of the results.
The advantage of this class approaches is that, a training machine can automatically 2 Predictive inference, which can forecast the results according determine a best Bayesian network model with structure and to the causes. This study between different reasons with the same result. An example of Bayesian networks is shown in Figure 1. According to the features of the shareholder value, economic profit and the key value drivers or the key risk indicators KRIs influencing economic profit are Figure 1.
Example of Bayesian Network [6] variables labeled on nodes. It can be obtained by either historical data or expert probability of n variables can be decomposed according to a judgment. This study focuses on data driven, i. The obtained results will be compared to statistical analysis methods. The model can be then used for enterprise value prediction and scenario analysis. WEKA offers various where is the probability of the joint event search algorithms.
This study chose TAN search algorithm, which has exhibited excellent performance in data mining [7]. The algorithm can produce a causal-effect graph, which is 4. In this paper, the external environments of expressway projects.
The external risk Bayesian Network model is applied for the scenario analysis of factors are uncontrollable and difficult to prevent. The internal enterprise risk management. In particular, 12, 13]. All variables or key value drivers key risk method for machine learning algorithms. The method starts with indicators from empirical exploration are shown in Table 1.
The assessment of This study used monthly quantitative non- financial data, part model performance will iterate k times. For the ith iteration, the of the financial perspective, amassed from financial report since ith subset is selected as the test dataset, and the remaining of fiscal year.
And, the key performance indicators subsets are merged into the training dataset used for model of operational activity are also used as the key value measures, construction. It is observed that each subset is used as a test obtained from the performance report since of fiscal dataset once.
Typically, the model performance is measured by year. However, due to small sample size 72 records , selection accuracy rate. The accuracy rate of the model using k-fold cross of relevant variables is suggested for developing a useful model.
In Variable selection is important on account of irrelevant and this study, fold cross validation was used to evaluate the redundant features may confuse the learning algorithm and model performance. Therefore, a small number of predictive variables are preferred. In this Bayesian learning can be used with discrete or continuous study, the variables are selected based upon the correlations and variable type.
Prior to model construction, the preprocessing of conditional correlations among variables. The variables that face is to decide the number of states for discretization. We have significant correlation are assumed to be related with the started from two states to five states and tested the model with economic profit or the value of economic profit depends on all samples. When continuous variables were discretized into 2 these variables. All variables or key risk factors influencing The findings showed that Variable Name Explanation discretizing continuous variables into 2 or 3 states led to the best Traffic x1 Traffic Volume per day performance.
It is Area income x3 Income from rental area observed that x9, x14, x15, and x16 disappear since they are disconnected nodes. BayesNet- D- Q weka. According to the Search. Remove- R,, Network is promising for identifying key risk factors used in 14,,weka. Instances: 72 Attributes: 9 revenue 5. Using the same eight variables, multiple regression yields accident the coefficient of determination R Square valued This indicates that Traffic volume per day Traffic , EP Time of travel drivetime , Cost of Route Maintenance Test mode: fold cross- validation routemaintencost , Income from rental area areaincome , Time taken to build model: 0.
Save to Library Save. Create Alert Alert. Share This Paper. Background Citations. Methods Citations. Results Citations. Topics from this paper. Bayesian network. Approximation algorithm. Citation Type. Has PDF. Publication Type. More Filters. Bayesian Networks: Representation and Inference. View 2 excerpts, cites background. On the combination of logical and probabilistic models for information analysis. Applied Intelligence. Highly Influenced.
View 5 excerpts, cites background and methods. Scalable Integration of Analytics and Visualization. View 1 excerpt, cites background. View 4 excerpts, cites background and methods.
Information Science and Statistics. Probabilistic networks, also known as Bayesian networks and influence diagrams, have become one of the most promising technologies in the area of applied artificial intelligence, offering intuitive, … Expand.
0コメント