Decision trees methodology

Understand decision tree methodologies used to evaluate supply chain design decisions under uncertainty. 

Like most supervised learning methods, the goal of the decision tree methodology is to develop classification rules that determine the class of any object from the values of the object s attributes. In the case of decision trees, as the name implies, the classification rules are embodied in a knowledge representation fonnalism called a decision tree. This method has been used to derive structure-activity relationships and to learn classification rules for reactions. 

A decision tree for Design Methodology is illustrated in Fig. 3.2. Each step in the tree is explained briefly below. The steps have also their own subtrees, which are described separately. 

The orthogonality of a set of molecular descriptors is a very desirable property. Classification methodologies such as CART (11) (or other decision-tree methods) are not invariant to rotations of the chemistry space. Such methods may encounter difficulties with correlated descriptors (e.g., production of larger decision trees). Often, correlated descriptors necessitate the use of principal components transforms that require a set of reference data for their estimation (at worst, the transforms depend only on the data at hand and, at best, they are trained once from some larger collection of compounds). In probabilistic methodologies, such as Binary QSAR (12), approximation of statistical independence is simplified when uncorrelated descriptors are used. In addition,... 

Only a small subset of method-specific questions are shown in Table 2.3 because of the breadth of chemometric and measurement methodology. Understanding the information in Chapters 3-5 will help generate chemometric-related questions. The decision trees in these chapters can also help determine appropriate method(s) for a given situation. In time, your experience will also become a valuable resource. 

Data mining is a new methodology for improving the quality and effectiveness of business and scientific decision-making processes.[76] There are currently several data mining techniques available. The decision-tree method is one of the most important techniques.[77] The attribute-splitting criteria of C5.0 decision tree is entropy, which is used to describe uncertainty of a stochastic experiment. X = pi(l), P2(2),..., pn n) is used to describe a stochastic experiment, in which p, denotes the probability of one experiment whose result is i. So the entropy function can be expressed as Equation (7.5) ... 

This paper has compared the following decision tree models for dependence assessment in HRA SPAR-H, the EPRI HRA Calculator , DEPEND-HRA, the HRA methodology used in the Surry Power Station. 

The next step we see in RTM is to introduce passive and active control which can account for part to part variability and reduce the scrap rate and make the process more reliable and attract manufacturers to consider lightweight composites as a serious alternative to stamped aluminum parts. There are control algorithms available which in unison with flow simulations (offline in which one anticipates the differences and creates a decision tree based on the disturbance that can be identified with sensors, and online simulations which try to correct the flow during injection) can be integrated in an automated environment which will make RTM competitive for high volume production. This methodology has been demonstrated in a laboratory environment and hopefully will be adopted by some industries to reap the benefits. Until then RTM will serve as prime candidate for low volume specific design components. 

FIGURE 15.1 Characterization of two hypothetical taxa using discriminant analysis (A) and decision trees (B). Although both variables in each group show a considerable overlap, both methodologies allow the identification of the limits of each group in the morphospace defined by the variables. 

Summarize the basic steps in the decision tree analysis methodology. 

Once acquired, there are many ways of representing the knowledge in the knowledge base, including production rules, frames, semantic networks, decision tables, and trees and objects. Probably the most common methodology is the production rule, which expresses the relationship between several pieces of information by way of conditional statements that specify sections under certain sets of conditions, for example ... 

The methodology of PSA is in principle a combination of event tree and fault tree analyses. The target of the analysis decisively determines the selection and the definition of modes in which the technical system is at the end of an event sequence. These modes comprise both the safe ones resulting from successful safety measures and also those unwanted ones characterized by a certain release of hazardous materials. A PSA is extended to a PRA, if the consequences of the different system modes are determined and linked with the event frequencies of those modes. 

Borysiewicz, M. et al., 2013. An application of the Value Tree Analysis methodology within an Integrated Risk Informed Decision Making on the nuclear facilities. Reliability Engineering and System Safety, in press. 

---