基于贝叶斯网络的工程项目进度完工概率分析

针对传统计划评审技术(Program Evaluation and Review Technique,PERT)在计算完工概率时假设条件的局限性(假设条件与工程实际存在偏差,导致完工概率偏大),提出了基于贝叶斯网络的施工进度完工概率分析方法.首先,分析了贝叶斯网络与进度计划网络之间的相似性,将两者结合起来构建了贝叶斯进度网络;在此基础上,综合考虑贝叶斯网络在节点取值及概率计算方面的优越性,并结合工程项目的不确定性及复杂性特点,建立了基于贝叶斯网络的施工进度完工概率分析模型.最后,将该模型应用于具体工程进行实例分析,验证了模型的可行性与有效性.研究结果表明:基于贝叶斯网络的进度完工概率模型充分考虑了工程施工中的风险因素,其结果能更客观地反映工程实际,可为工程项目决策者提供可靠的依据.     

A discrete event simulation model in the case of managing a software project

The design and development of large-scale software projects is a complex endeavor, often facing problems like cost and schedule overruns as well as low quality. Over the last years the management of software development projects has been recognized as the cornerstone point of seeking improvement and solutions. Simulation modeling of the software project process is gaining interest among academics and practitioners, as a method to tackle the complex questions with which relevant enterprises are confronted. It offers support on several issues, such as defining software product development strategies, decision-making regarding process improvement and training, in a time span ranging from a short portion of the life cycle to long term product evolution, with organization-wide implications. The aim of this work is to implement a model simulating a core part of a software project process, enabling the estimation of several project development details such as delivery times and quality metrics. The purpose of the model is to assist project managers in control and monitoring, but also in identifying the best planning alternatives. The model scope covers a portion of the life cycle of an incremental software development venture.     

Systemic risk assessment: a case study

Project Risk Registers have been used extensively for many years. However, they do not account for the interaction between risks, for example, the occurrence of one risk exacerbating other risks or portfolios of risks being more significant than the sum of the individual risks. This leads to the need to consider ‘risk systemicity’ as a part of risk analysis. This paper reports on a specific case for a large multinational project based organization, one that the authors had been involved with in the analysis of a number of projects that had massive cost overruns. Following these analyses the organization was persuaded of the importance of risk systemicity. The organization therefore engaged the authors to develop a ‘Risk Filter’. This filter is a tool for identifying areas of risk exposure on future projects and creating a framework for their investigation. The ‘Risk Filter’ is now used on all projects ever since its introduction; by the end of May 2003 it had been used by nine divisions, on over 60 major projects, and completed by 450 respondents. It is also used at several stages during the life of a project to aid in the risk assessment and management of each project, and contributes to a project database.     

Project cost dynamics for development policy-making

Cost and time overruns have been a common characteristic of development projects in many countries. This paper presents a theory of cost and time overruns based on project cost structure. The cost of a development project consists of base cost and progress cost. Base cost keeps a project ready for physical progress. Progress cost creates real physical progress on the project. This cost structure has an important inherent dynamic characteristic with implications for the efficiency and effectiveness of project management. An imbalance between annual budget and ongoing projects results in an increasing inefficiency and ineffectiveness unrelated to the quality of management of individual projects. The policy governing the starting rate of the new projects becomes very important in improving project performances. The paper also suggests some policy recommendations for commencing new projects.     

Parameter identification for rubber materials with artificial higher dimensional data

In general, laboratory test render only a limited number of experimental data. Consequently, the prediction of material behaviour becomes a difficult task and, moreover, a statistical analysis with a statistically based approach is almost impossible. As a remedy to increase the number of data, artificial data are generated by stochastic simulation. As a consequence an arbitrary number of data is available and the process of parameter identification can be analysed statistically. Here, the special challenge is the consideration of spatial and inhomogeneous problems. In this work artificial data are generated for a elastomer strip with hole under tension. The inhomogeneous stress/strain fields are optically measured with an Aramis/GOM system and have to be fitted to a stochastic model in order to generate artificial data. B-Splines are applied to fit the geometry of the test specimen and the measured data in space as well as in time. Parameter identifications applied and the resulting material parameters are statistically analysed. In the example, a statistical analysis of an Ogden model is performed. (© 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim)     

Spectral asymptotics for Kac–Murdock–Szegő matrices

Szeg?’s First Limit Theorem provides the limiting statistical distribution of the eigenvalues of large Toeplitz matrices. Szeg?’s Second (or Strong) Limit Theorem for Toeplitz matrices gives a second order correction to the First Limit Theorem, and allows one to calculate asymptotics for the determinants of large Toeplitz matrices. In this paper we survey results extending the First and Second Limit Theorems to Kac–Murdock–Szeg? (KMS) matrices. These are matrices whose entries along the diagonals are not necessarily constants, but modeled by functions. We clarify and extend some existing results, and explain some apparently contradictory results in the literature.     

Asymptotically efficient parameter estimation for ordinary differential equations

Parameter estimation for ordinary differential equations arises in many fields of science and engineering. To be the best of our knowledge, traditional methods are often either computationally intensive or inaccurate for statistical inference. Ramsay et al. (2007) proposed a generalized profiling procedure. It is easily implementable and has been demonstrated to have encouraging numerical performance. However, little is known about statistical properties of this procedure. In this paper, we provide a theoretical justification of the generalized profiling procedure. Under some regularity conditions, the procedure is shown to be consistent for a broad range of tuning parameters. When the tuning parameters are sufficiently large, the procedure can be further shown to be asymptotically normal and efficient.     

On the inconsistency of Bayesian non-parametric estimators in competing risks/multiple decrement models

In the competing risks/multiple decrement model, the joint distribution is often not identifiable given only the observed time of failure and the cause of failure. The traditional approach is consequently to assume a parametric model. In this paper we shall not do this, but rather assume a Bayesian stance, take a Dirichlet process as a prior distribution, and then calculate the posterior distribution given the data. In this paper we show that in dimensions ? 2, the posterior mean yields an inconsistent estimator of the joint probability law, contrary to the common assumption that the prior law ‘washes out’ with large samples. For single decrement mortality tables however, the non-parametric Bayesian method allows a flexible method for adjusting a standard mortality table to reflect mortality experience, or covariate information.     

Structural reliability and Monte Carlo simulation

Using elementary engineering mechanics, a simple boom structure is analysed and its reliability is assessed. It is assumed that four of the structural variables are random with a known joint probability distribution function. From the structural analysis, an explicit expression for failure is obtained. The reliability of the structure is computed analytically from first principles of probability theory. Additionally, it is shown that Monte-Carlo simulation can be implemented more readily with results that compare favourably to the theoretical calculations. Thus, the power and utility of Monte-Carlo simulation are demonstrated. Specifically, the boom is analysed for different probability distribution functions for the underlying components. The analytical and numerical analysis contained in this application are appropriate for any upper level undergraduate course in applied mathematics, scientific computation or engineering reliability where the Monte-Carlo method is being studied.     

Parameter identification for stochastic hybrid systems using randomized optimization: A case study on subtilin production by Bacillus subtilis

In this paper we study the parameter identification problem for a stochastic hybrid model of the production of the antibiotic subtilin by the bacterium B. subtilis. We pursue a simulation-based approach, in which the fit of candidate parameter values is evaluated by comparing simulated model trajectories with experimental data. Several score functions are considered to capture the goodness of the fit. Parameter estimation is accomplished via an evolutionary strategy that iteratively selects the best fitting parameters. Identifiability issues are discussed and are explored numerically by a Markov Chain Monte Carlo approach.     

The range inter‐event process in a symmetric birth–death random walk

This paper provides new results for the range inter‐events process of a birth–death random walk. Motivations for determining and using the inter‐range event distribution have two sources. First, the analytical results we obtain are simpler than the range process and make it easier, therefore, to use statistics based on the inter‐range event process. Further, most of the results for the range process are based on long‐run statistical properties which limits their practical usefulness while inter‐range events are by their nature ‘short‐term’ statistics. Second, in many cases, data on amplitude change are easier to obtain and calculate than range and standard deviation processes. As a results, the predicted statistical properties of the inter‐range event process can provide an analytical foundation for the development of statistical tests that may be used practically. Application to outlier detection, volatility and time‐series analysis is discussed. Copyright © 2001 John Wiley & Sons, Ltd.     

Defending and improving the ‘slotting fee’: how it can benefit all the stakeholders dealing with a newsvendor product with price and effort-dependent demand

‘Slotting fee’ (hereafter ‘SF’) is an upfront fee a ‘supplier’ is required to pay a retailer in order to have his product sold on the retailer's shelves. It is becoming increasingly common, but also widely reviled. This paper considers a newsvendor product whose expected demand is dependent on retail price and sales effort. The question we pose is: given that the Stackelberg-dominant retailer has to choose a pricing contract with which she transacts with the supplier, how would the supply-chain stakeholders fare when the retailer implements SF instead of another practical pricing contract? We show that, contradicting its negative public image, SF empowers the dominant retailer to specify contract terms that will benefit all the stakeholder-groups. That is, the supplier's and the retailer's profits are higher, the production workers are asked to produce more, and the consumers pay a lower retail price. We also propose a new ‘composite’ contract format that incorporates both the SF and ‘buyback’ features. This composite format empowers the retailer to provide even greater benefits to the supply-chain's stakeholders.

     

Analysis of third-party warehousing contracts with commitments

This paper considers multi-period warehousing contracts under random space demand. A typical contract is specified by a starting space commitment plus a certain number of times at which the commitment can be further modified. Three forms of contracts are analysed: (1) there is a restriction on the range of commitment changes and the schedule for the changes is preset by the warehouser; (2) the same as form (1) but there is no restriction on the range; (3) the same as form (2) but the schedule for the changes is chosen by the user. We explore properties and algorithms for the three problems from the user's perspective. Solutions of simple form are obtained for the first two models and an efficient dynamic programming (DP) procedure is proposed for the last. A numerical comparison of the total expected leasing costs suggests that under certain demand patterns, contract forms (2) and (3) could be cost effective.     

A Nonequilibrium Statistical Model of Spectrally Truncated Burgers‐Hopf Dynamics

Exact spectral truncations of the unforced, inviscid Burgers‐Hopf equation are Hamiltonian systems with many degrees of freedom that exhibit intrinsic stochasticity and coherent scaling behavior. For this reason recent studies have employed these systems as prototypes to test stochastic mode reduction strategies. In the present paper the Burgers‐Hopf dynamics truncated to n Fourier modes is treated by a new statistical model reduction technique, and a closed system of evolution equations for the mean values of the m lowest modes is derived for m ? n. In the reduced model the m‐mode macrostates are associated with trial probability densities on the phase space of the n‐mode microstates, and a cost functional is introduced to quantify the lack of fit of paths of these densities to the Liouville equation. The best‐fit macrodynamics is obtained by minimizing the cost functional over paths, and the equations governing the closure are then derived from Hamilton‐Jacobi theory. The resulting reduced equations have a fractional diffusion and modified nonlinear interactions, and the explicit form of both are determined up to a single closure parameter. The accuracy and range of validity of this nonequilibrium closure is assessed by comparison against direct numerical simulations of statistical ensembles, and the predicted behavior is found to be well represented by the reduced equations. © 2014 Wiley Periodicals, Inc.     

The effect of activity-based instruction on conceptual development of seventh grade students in probability

The purpose of this study is to investigate and compare the effects of activity-based and traditional instructions on students’ conceptual development of certain probability concepts. The study was conducted using a pretest–posttest control group design with 80 seventh graders. A developed ‘Conceptual Development Test’ comprising 12 open-ended questions was administered on both groups of students before and after the intervention. The data were analysed using analysis of covariance, with the pretest as covariate. The results revealed that activity-based instruction (ABI) outperformed the traditional counterpart in the development of probability concepts. Furthermore, ABI was found to contribute students’ conceptual development of the concept of ‘Probability of an Event’ the most, whereas to the concept of ‘Sample Space’ the least. As a consequence, it can be deduced that the designed instructional process was effective in the instruction of probability concepts.     

不同先验分布下的后验分布确定土力学参数

岩土工程中各土层参数的取值是根据现场及室内试验数据,采用经典统计学方法进行确定的,但这往往忽略了先验信息的作用。与经典统计学方法不同的是,Bayes法能从考虑先验分布的角度结合样本分布去推导后验分布,为岩土参数的取值提供一种新的分析方法。岩土工程勘察可视为对总体地层的随机抽样,当抽样完成时,样本分布密度函数是确定的,故Bayes法中的后验分布取决于先验分布,因此推导出两套不同的先验分布:利用先验信息确定先验分布及共轭先验分布。通过对先验及后验分布中超参数的计算,当样本总体符合N(μ,σ2)正态分布时,对所要研究的未知参数μ和σ展开分析,综合对比不同先验分布下后验分布的区间长度,给出岩土参数Bayes推断中最佳后验分布所要选择的先验分布。结果表明:共轭情况下的后验分布总是比无信息情况下的后验区间短,概率密度函数分布更集中,取值更方便。在正态总体情形下,根据未知参数μ和σ的联合后验分布求极值方法,确定样本总体中最大概率均值μ_max和方差σ_max作为工程设计采用值,为岩土参数取值方法提供了一条新的路径,有较好的工程意义。     

大工程决策中公众意愿与地方政府行为偏好研究

在公共事务决策中公众参与正影响着地方政府决策。基于博弈视角在地方政府与公众的策略互动中,博弈收敛速度的提高能够降低双方的成本从而提高社会效率。建立基于大工程议案决策的地方政府与公众动态博弈过程,应用马尔可夫决策研究博弈收敛的条件并分析了基于折扣准则下地方政府采取妥协策略时的最优停止条件。讨论了地方政府对不稳定的容忍度、地方政府选择妥协的概率与博弈收敛速度三者之间的关系。研究表明大工程预期经济收益指标与地方政府对地方不稳定的容忍度呈正向关系,地方政府对地方不稳定的容忍度与双方关于大工程议案决策博弈的收敛速度呈反向关系,且公众支持度对地方容忍度的影响依概率分布变化。     

Managing Cost Overrun Risk in Project Funding Allocation

This paper discusses decision making of project funding allocation under uncertain project costs. Because project costs are uncertain and funding allocations may not necessarily match the costs required, each project is inherently subject to a cost overrun risk (COR). In this paper, a model is proposed in which project cost is treated as a factor with a probability density function. The decision maker then allocates the total funding to the projects while minimizing a weighted sum of mean and variance of the COR of the project portfolio. Some properties of project COR are derived and interpreted. Optimal funding allocation, in relationship to factors such as various project sizes and riskiness, project interdependency, and the decision maker’s risk preference, is analyzed. The proposed funding allocation model can be integrated with project selection decision-making and provides a basis for more effective project control.     

An Unified Computational Framework for Parameter Identification of Material Models in Finite Inelasticity

Parameter identification processes concern the determination of parameters in a material model in order to fit experimental data. We provide a distinct, unified algorithmic setting of a generic class of material models and discuss the associated gradient–based optimization problem. Gradient–based optimization algorithms need derivatives of the objective function with respect to the material parameter vector κ . In order to obtain the necessary derivatives, an analytical sensitivity analysis is pointed out for the unified class of algorithmic material models. The quality of the parameter identification is demonstrated for a representative example.