Warning Decision-Making for Landslide Dam Breaching Flood Using Influence Diagrams

Influence diagrams are a powerful graphic representation for decision models, complementary to decision trees. Influence diagrams and decision trees are different graphic representations for the same underlying mathematical model and operations. This article describes the elements of an influence diagram, and shows several familiar decision problems represented as decision trees and as influence diagrams. The authors also contrast the information highlighted in each graphic representation, demonstrate how to calculate the expected utilities of decision alternatives modeled with an influence diagram, provide an overview of the conceptual basis of the solution algorithms that have been developed for influence diagrams, discuss the strengths and limitations of influence diagrams relative to decision trees, and describe the mathematical operations that are used to evaluate both decision trees and influence diagrams. They use clinical examples to illustrate the mathematical operations of the influence-diagram-evaluation algorithm; these operations are arc reversal, chance node removal by averaging, and decision node removal by policy determination. Influence diagrams may be helpful when problems have a high degree of conditional independence, when large models are needed, when communication of the probabilistic relationships is important, or when the analysis requires extensive Bayesian updating. The choice of graphic representation should be governed by convenience, and will depend on the problem being analyzed, on the experience of the analyst, and on the background of the consumers of the analysis.

An Ms 6.5 earthquake shocked the Ludian County, Yunnan Province, China, on 3 August 2014 and triggered the Hongshiyan landslide dam. The dam, with a height of 83 m and a lake capacity of 260 × 106 m3, threatened more than 10,000 people. A unique feature of this landslide dam was that it formed between a man-made dam and a hydropower plant. An existing drainage tunnel connecting the lake and the hydropower plant became a natural drainage conduit for the landslide dam, which played an important role in the mitigation of the landslide dam risks. This paper reports a quantitative risk assessment for the Hongshiyan landslide dam considering both engineering and non-engineering risk mitigation measures. The risk assessment is divided into three stages according to the implementation of two engineering measures: construction of a diversion channel and excavation of a branch drainage tunnel. The dam breaching hydrographs, flood zones, population at risk, and likely fatalities in each of the three stages are analysed. The optimum evacuation strategy in each stage is also studied based on the principle of minimum total consequence. It is found that the diversion channel decreases the dam breaching peak discharge and the associated risks significantly. The branch drainage tunnel prevent the landslide dam from overtopping failure in non-flooded period; however, the landslide dam may fail by overtopping in a future flood if the inflow rate is larger than the outflow rate through the drainage tunnels, resulting in serious losses of lives and properties. The dam breaching risks in all the three stages could be largely reduced by the optimal evacuation decision, which shows that timely evacuation is vital to save life and properties. The study provides a scientific basis for decision making in landslide dam risk management.

This article, written by Senior Technology Editor Dennis Denney, contains highlights of paper SPE 127761, ’Probabilistic Modeling for Decision Support in Integrated Operations,’ by Martin Giese, University of Oslo, and Reidar B. Bratvold, SPE, University of Stavanger, prepared for the 2010 SPE Intelligent Energy Conference and Exhibition, Utrecht, Netherlands, 23-25 March. The paper has not been peer reviewed. A system was designed to assist operational decisions on the basis of real-time sensor readings in a typical scenario: While drilling close to the transition to a high-pressure formation, gas influx is observed. The technology used Bayesian networks and influence diagrams (IDs) to apply the method of decision analysis. The resulting ID was tested with a simulation. Introduction An assumption of integrated operations is that all information provided by the technology will help to make better decisions faster. But for this to happen, information must be presented in an accessible, useful, and decision-relevant format. Decision-analysis tools provide decision makers with logical and consistent recommendations based on the available data. This decision-analysis method was applied to a real case, representative of a typical scenario in drilling operations. The tool enabled good decisions consistently. However, effort is required and challenges exist in the modeling process along with problems regarding the way decision making is treated in the industry. Modeling and Solving A decision tree provides a graphical illustration of all the uncertainties, decisions, and payoffs associated with a decision situation. Although application of decision trees is intuitive for simple problems, the method loses its appeal for complex decisions. A symmetric decision tree grows exponentially with the number of variables in the decision domain, which often makes its use impractical. For complex decisions, the ID can provide a level of insight and transparency that cannot be achieved with the equivalent decision tree. The ID is a compact graphical-model representation for reasoning under uncertainty and was developed to assist structuring and analysis of complex decisions. The ID has two advantages—structuring and probabilistic evaluation. The structuring provides a graphical representation of the inter-relationships among the decision-basis elements (i.e., alternatives, information, and preferences), while the evaluation appraises the resulting decision structure and provides an assessment of its expected utility given the quantified decision basis. The representational framework of an ID captures the probabilistic relationships among decisions (alternatives), key uncertainties (information), and utilities (values or preferences). The ID model comprises four types of nodes. Decision nodes specify a set of alternatives. Chance nodes represent the key uncertainties in the decision domain. Deterministic nodes represent either constants or intermediate calculations. Value nodes represent the criteria used to make the decisions.

Numerical simulation of landslide dam overtopping failure considering headward erosion

Abstract. Tangjiashan landslide dam, which was triggered by the Ms = 8.0 Wenchuan earthquake in 2008 in China, threatened 1.2 million people downstream of the dam. All people in Beichuan Town 3.5 km downstream of the dam and 197 thousand people in Mianyang City 85 km downstream of the dam were evacuated 10 days before the breaching of the dam. Making such an important decision under uncertainty was difficult. This paper applied a dynamic decision-making framework for dam-break emergency management (DYDEM) to help rational decision in the emergency management of the Tangjiashan landslide dam. Three stages are identified with different levels of hydrological, geological and social-economic information along the timeline of the landslide dam failure event. The probability of dam failure is taken as a time series. The dam breaching parameters are predicted with a set of empirical models in stage 1 when no soil property information is known, and a physical model in stages 2 and 3 when knowledge of soil properties has been obtained. The flood routing downstream of the dam in these three stages is analyzed to evaluate the population at risk (PAR). The flood consequences, including evacuation costs, flood damage and monetized loss of life, are evaluated as functions of warning time using a human risk analysis model based on Bayesian networks. Finally, dynamic decision analysis is conducted to find the optimal time to evacuate the population at risk with minimum total loss in each of these three stages.

TWO ALGORITHMS THAT LEARN DNF BY DISCOVERING RELEVANT FEATURES

Chapter 5 - Decision Analysis Fundamentals

Decades of research have repeatedly shown that people perform poorly at estimating and understanding conditional probabilities that are inherent in Bayesian reasoning problems. Yet in the medical domain, both physicians and patients make daily, life-critical judgments based on conditional probability. Although there have been a number of attempts to develop more effective ways to facilitate Bayesian reasoning, reports of these findings tend to be inconsistent and sometimes even contradictory. For instance, the reported accuracies for individuals being able to correctly estimate conditional probability range from 6% to 62%. In this work, we show that problem representation can significantly affect accuracies. By controlling the amount of information presented to the user, we demonstrate how text and visualization designs can increase overall accuracies to as high as 77%. Additionally, we found that for users with high spatial ability, our designs can further improve their accuracies to as high as 100%. By and large, our findings provide explanations for the inconsistent reports on accuracy in Bayesian reasoning tasks and show a significant improvement over existing methods. We believe that these findings can have immediate impact on risk communication in health-related fields.

Single‐objective‐based decision‐tree analysis has been extensively and successfully used in numerous decision‐making problems since its formal introduction by Howard Raiffa more than two decades ago. This paper extends the traditional methodology to incorporate multiple noncommensurate objective functions and use of the conditional expected value of the risk of extreme and catastrophic events. The proposed methodology considers the cases where (a) a finite number of actions are available at each decision node and (b) discrete or continuous states of nature can be presented at each chance node. The proposed extension of decision‐tree analysis is introduced through an example problem that leads the reader step‐by‐step into the methodological procedure. The example problem builds on flood warning systems. Two noncommensurate objectives—the loss of lives and the loss of property (including monetary costs of the flood warning system)–are incorporated into the decision tree. In addition, two risk measures—the common expected value and the conditional expected value of extreme and catastrophic events—are quantified and are also incorporated into the decision‐making process. Theoretical difficulties associated with the stage‐wise calculation of conditional expected values are identified and certain simplifying assumptions are made for computational tractibility. In particular, it is revealed that decisions concerning experimentation have a very interesting impact on the noninferior solution set of options—a phenomenon that has no equivalence in the single‐objective case.

Outburst floods caused by breaches of landslide dams may cause serious damages and loss of lives in downstream areas; for this reason the study of the dynamic of the process is of particular interest for hazard and risk assessment. In this paper we report a field-scale landslide dam failure experiment conducted in Nantou County, in the central of Taiwan. The seismic signal generated during the dam failure was monitored using a broadband seismometer and the signal was used to study the dam failure process. We used the short-time Fourier transform (STFT) to obtain the time–frequency characteristics of the signal and analyzed the correlation between the power spectrum density (PSD) of the signal and the water level. The results indicate that the seismic signal generated during the process consisted of three components: a low-frequency band (0–1.5 Hz), an intermediate-frequency band (1.5–10 Hz) and a high-frequency band (10–45 Hz). We obtained the characteristics of each frequency band and the variations of the signal in various stages of the landslide dam failure process. We determined the cause for the signal changes in each frequency band and its relationship with the dam failure process. The PSD sediment flux estimation model was used to interpret the causes of variations in the signal energy before the dam failure and the clockwise hysteresis during the failure. Our results show that the seismic signal reflects the physical characteristics of the landslide dam failure process. The method and equipment used in this study may be used to monitor landslide dams and providing early warnings for dam failures.

Conversion of a colorectal cancer guideline into clinical decision trees with assessment of validity

Motivation and Objectives Neuroblastoma (NB) is the most common pediatric solid tumor characterized by clinical and molecular risk factors. The mortality is about fifty percent and this makes exploration of new and more effective risk factors for improving stratification mandatory. Hypoxia is a condition of low oxygen tension occurring in poorly vascularized areas of the tumor associated with poor prognosis. We had previously defined a robust gene expression signature measuring the hypoxic component of NB tumors (NB-hypo) that is a novel, independent risk factor (Fardin et al., 2010). Integrating classical risk factors with NB-hypo could improve the stratification of NB patients. We wanted to develop a prognostic classifier of NB patients’ outcome blending existing knowledge on clinical and molecular risk factors with the prognostic NB-hypo signature. Furthermore, we were interested in the decision tree classifier that outputs explicit rules easily translated into the clinical setting. Methods A total of 182 NB patients were enrolled on the bases of availability of gene expression profile by Affymetrix GeneChip HG-U133plus2.0, clinical and molecular information. NB tumor stage was defined according to the International NB Staging System (INSS). Age at diagnosis was dichotomized as greater or equal than 1 year and less than one year. MYCN status was amplified or normal. Good and poor outcome were defined as the patient’s status alive or dead 5 years after diagnosis respectively. The risk group was assigned according to the International Neuroblastoma Risk Group (INRG) Consensus Pretreatment Classification Schema. The 182 NB patients cohort was clustered in High and low hypoxia by k-means analysis of the 62 probsets constituting the NB-hypo signature previously described to measure tumor hypoxia (Fardin et al., 2009). We utilized the k-means algorithm implemented in the WEKA software (Hall et al., 2009) setting up number of clusters to 2, 500 iterations, preserving instances order and using Manhattan distance. The classification was performed by induction of decision trees. We utilized the Weka J48 implementation of the popular C4.5 algorithm (Kotsiantis, 2007; Murthy, 1998) and we set up the following options: pruning parameter was 0.25, pruning method was sub-tree raising and minimum number of instances per leaf was 2. Each leaf of the decision tree classifier identifies a non overlapping group of patients and each decision node identifies a branch which splits the dataset. We utilized Fisher’s exact test to measure the statistical significance of groups and branches. Fisher’s test was utilized in the context of decision trees to design a top-down approach to prune out non statistically significant branches (Liu et al., 2010). For each leaf we counted the number of correctly classified (named n) and the number of incorrectly classified instances (named m). We considered the marginal totals y and ¬y which represent poor and good outcome patients respectively. We designed a 2x2 contingency table of the two possible outcomes (Good or Poor) against the number of instances included in a give leaf and the remaining instances. Application of the Fisher exact test to this table generates a p value giving the probability of observing 2x2 table, or more extreme tables, knowing the marginal totals (y, ¬y) and assuming independency among the patients in a specific leaf and those in other leaves. Results and Discussion Patients were stratified in good and poor outcome on the bases of the following risk factors: Age at diagnosis, INSS stage, MYCN status and NB-hypo. The algorithm generated a decision tree classifier composed by 3 decision nodes and 7 leaves covering 87% of non overlapping good outcome patients and 100% of non overlapping poor outcome patients. Each path from the root to a leaf utilizes some, but not all, considered risk factors. Interestingly NB-hypo was included in the decision node that stratified stage 3 tumors demonstrating its usefulness in NB patients’ stratification. The leaves classifying good outcome patients had the very low error of 2% indicating a good performance of the algorithm in predicting this class. In contrast, the classification of poor outcome patients produced the leaf with the highest error of 13%. This leaf includes stage 4 tumors that are traditionally difficult to stratify by any known risk factor. To test statistical significance of the splits performed by the algorithm and the groups of patients, we utilized the Fisher’s exact test with a confidence set at 95%. The results showed that the groups obtained at each split were statistically significant. Furthermore, analysis of single groups of patients identified by leaves demonstrated that some, but no all reached the significance threshold. The significant leaves included the NB-hypo among the utilized risk factors. These results further strengthen the value of NB-hypo in predicting patients’ outcome. Lack of significance was often associated with a rather low number of patients in the leaf. This is a limitation of the divide-and-conquer approach of the algorithm, applied to relatively small patients’ cohorts, that recurrently splits the dataset. We then assessed the concordance of the predictions with INRG risk assessment. High Risk patients were correctly included in the leaves classifying poor outcome patients and Low Risk patients mapped correctly in the good outcome leaves. Interestingly, NB-hypo generated a leaf identifying a new group of poor outcome patients, sharing the high NB-hypo, whose characteristic fell into both High and Intermediate Risk. We collected and analyzed the results of 1000 10-fold cross validations and we observed that most classifiers had 7 leaves and only 5 out of 10^4 deviated from this pattern. The recurrence of 7 leaves demonstrated the high stability of the decision tree classifier that we generated. Analysis of the pruning parameters revealed optimal performance in the range of 0.1-0.3 in line with what used in this study. The path to reach each leaf can be easily transposed into a “if…than…” rule that, in turn provides an easy readout of the classifier, precious for translating the classification into the clinical setting. In conclusion, we demonstrated that the decision tree algorithm C4.5 can derive explicit rules for NB patients stratification if classical risk factors are blended with the NB hypo signature. These rules are statistically significant and quite stable and suitable to be conveyed to the clinic to design new therapies perhaps taking hypoxia into consideration as a potential target. Acknowledgements The work was supported by the Fondazione Italiana per la Lotta al Neuroblastoma, the Associazione Italiana per la Ricerca sul Cancro, the Società Italiana Glicogenosi, the Fondazione Umberto Veronesi and the Ministero della Salute Italiano. Davide Cangelosi and Fabiola Blengio are recipients of a fellowship from the Fondazione Italiana per la Lotta al Neuroblastoma. References Fardin P, Barla A, Mosci S, Rosasco L, Verri A, Varesio L. (2009) The l1-l2 regularization framework unmasks the hypoxia signature hidden in the transcriptome of a set of heterogeneous neuroblastoma cell lines. BMC Genomics, 10:474. doi:10.1186/1471-2164-10-474 Fardin P, Barla A, Mosci S, Rosasco L, Verri A, et al. (2010) A biology-driven approach identifies the hypoxia gene signature as a predictor of the outcome of neuroblastoma patients, Journal of Molecular Cancer 9:1. 185. doi:10.1186/1476-4598-9-185 Hall M, Eibe F, Holmes G, Pfahringer B, Reutemann P, Witten IH. (2009) The WEKA Data Mining Software: An Update: SIGKDD Explorations. doi: 10.1145/1656274.1656278 Kotsiantis S.B. (2007) Supervised Machine Learning: A Review of Classification Techniques. Informatica 31:249-268. Liu W, Chawla S, Cieslak D, Chawla N. (2010) A Robust Decision Tree Algorithm for Imbalanced Data Sets. In: SDM. 766-777. Murthy S.K. (1998) Automatic Construction of Decision Trees from Data: A Multi-Disciplinary Survey, Data Mining Knowledge Discovery 2:4. 345-389. doi: 10.1023/A:1009744630224

Computing the confidence in a medical decision obtained from an influence diagram

Geomorphic hazard assessment of landslide dams in South Westland, New Zealand: fundamental problems and approaches

It is well recognized that the probability of fire start and conditional probability of fire flashover, given fire starts, may be expected in a given building within a broad range, depending on the applied fire safety system. Within the fire safety system the scale of active fire protection can be used to determine the design fire load. Similarly the probability of structural failure and extent of possible injuries from a fire depend considerably on the system of structural protection and arrangement of escape routes. Thus, risk assessment of structures under fire should be based on probabilistic concepts, taking into account various conditional probabilities, for example the proportion of fully developed fires of all started fires, i.e. the conditional probability that of fire flashover, provided that fire has started.It appears that Bayesian causal (belief) networks, supplemented by appropriate input data, may provide an effective tool to analyse the significance of various characteristics of a fire safety system in terms of the resulting probability of flash over and other unfavorable consequences. Moreover, Bayesian networks, supplemented by decision nodes and a number of utility nodes (influence diagram), make it possible to estimate the expected total risk for both the buildings and their occupants and the actions due to the fire.This review paper indicates that application of Bayesian causal (belief) networks may improve assessment of expected total risk, taking into account significant fire safety measures, including structural protection and arrangements of escape routes.