Dynamic systems identification with Gaussian processes

This paper describes the identification of nonlinear dynamic systems with a Gaussian process (GP) prior model. This model is an example of the use of a probabilistic non-parametric modelling approach. GPs are flexible models capable of modelling complex nonlinear systems. Also, an attractive feature of this model is that the variance associated with the model response is readily obtained, and it can be used to highlight areas of the input space where prediction quality is poor, owing to the lack of data or complexity (high variance). We illustrate the GP modelling technique on a simulated example of a nonlinear system.     

Simulation modelling in healthcare: reviewing legacies and investigating futures

This article proposes considered futures for the use of simulation as a problem solving technique within healthcare settings. Using a synthesis of trends identified by a selection of experts in the field, academics and industrialists, critical analysis was applied to find the differences between what exists and what could be created based on outlining some major themes. The survey data reveals that most respondents agree on following whole system approaches with more joined up modelling or mixed methods to tackle problems rather than single-solution-based practices. The article then presents options for how simulation could be used within the healthcare domain. Such options or futures could assist in identifying the critical barriers towards having a successful strategy and provide the basis for debate that will be necessary to attain it.     

Expert and novice facilitated modelling: A case of a Viable System Model workshop in a local food network

This paper provides an empirical study based on action research in which expert and novice facilitators in facilitated modelling workshops are compared. There is limited empirical research analysing the differences between expert and novice facilitators. Aiming to address this gap we study the behaviour of one expert and two novice facilitators during a Viable System Model workshop. The findings suggest common facilitation patterns in the behaviour of experts and novices. This contrasts literature claiming that experts and novices behave and use their available knowledge differently, and empirically supports the claim that facilitation skills can be taught to participants to enable them to self-facilitate workshops. Differences were also found, which led to the introduction of a new dimension—‘internal versus external’ facilitation. The implications of our findings for effective training and facilitation strategies in contexts in which external, expert facilitation is not always possible are also discussed, and limitations of this study are provided.     

Model development in discrete-event simulation and system dynamics: An empirical study of expert modellers

An empirical study comparing the model development process followed by experts in discrete-event simulation (DES) and system dynamics (SD) modelling is undertaken. verbal protocol analysis (VPA) is used to study the modelling process followed by ten expert modellers (5 SD and 5 DES). Participants are asked to build simulation models based on a case study and to think aloud while modelling. The generated verbal protocols are divided into seven modelling topics: problem structuring, conceptual modelling, data inputs, model coding, verification & validation, results & experimentation and implementation and then analyzed. Our results suggest that all modellers switch between modelling topics, however DES modellers follow a more linear progression. DES modellers focus significantly more on model coding and verification & validation, whereas SD modellers on conceptual modelling. Observations are made revealing some interesting differences in the way the two groups of modellers tackle the case. This paper contributes towards the comparison of DES and SD.     

Inference of the stochastic MAPK pathway by modified diffusion bridge method

The MAPK pathway is one of the well-known systems in oncogene researches of eukaryotes due to its important role in cell life. In this study, we perform the parameter estimation of a realistic MAPK system by using western blotting data. In inference, we use the modified diffusion bridge algorithm with data augmentation technique by modelling the realistically complex system via the Euler–Maruyama approximation. This approximation, which is the discretized version of the diffusion model, can be seen as an alternative OR approach with respect to the (hidden) Markov chain method in stochastic modelling of the biochemical systems where the data can be fully or partially observed and the time-course measurements are though to be collected at small time steps. Hereby, the modified diffusion bridge technique, which is based on the Markov Chain Monte Carlo (MCMC) methods, enables us to accurately estimate the model parameters, presented as the stochastic reaction rate constants, of the diffusion model under high dimensional systems despite loss in computational demand. In the estimation of the parameters, due to the complexity in the decision-making problems of the MCMC updates at different stages, we face with the dependency challenges. We unravel them by checking the singularity of the system in every stage of updates. In modelling, we also assume with/without-measurement error approaches in all states. But in order to evaluate the performance of both models, we initially implement them in a toy system. From the results, we observe that the model with measurement error performs better than the model without measurement error in terms of the mixing features of the MCMC runs and the accuracy of estimates, thereby, it is used for the parameter estimation of the realistic MAPK pathway. From the outcomes, we consider that the suggested approach can be seen as a promising alternative method in inference of parameters via different OR techniques in system biology.     

Nonparametric Bayesian topic modelling with the hierarchical Pitman–Yor processes

The Dirichlet process and its extension, the Pitman–Yor process, are stochastic processes that take probability distributions as a parameter. These processes can be stacked up to form a hierarchical nonparametric Bayesian model. In this article, we present efficient methods for the use of these processes in this hierarchical context, and apply them to latent variable models for text analytics. In particular, we propose a general framework for designing these Bayesian models, which are called topic models in the computer science community. We then propose a specific nonparametric Bayesian topic model for modelling text from social media. We focus on tweets (posts on Twitter) in this article due to their ease of access. We find that our nonparametric model performs better than existing parametric models in both goodness of fit and real world applications.     

An attribute weight based feedback model for multiple attributive group decision analysis problems with group consensus requirements in evidential reasoning context

In an evidential reasoning context, a group consensus (GC) based approach can model multiple attributive group decision analysis problems with GC requirements. The predefined GC is reached through several rounds of group analysis and discussion (GAD) in the approach. However, the GAD with no guidance may not be the most appropriate way to reach the predefined GC because several rounds of GAD will spend a lot of time of all experts and yet cannot help them to effectively emphasize on the assessments which primarily damage the GC. In this paper, an attribute weight based feedback model is constructed to effectively identify the assessments primarily damaging the GC and accelerate the GC convergence. Considering important attributes with the weights more than or at least equal to the mean of the weights of all attributes, the feedback model constructs identification rules to identify the assessments damaging the GC for the experts to renew. In addition, a suggestion rule is introduced to generate appropriate recommendations for the experts to renew their identified assessments. The identification rules are constructed at three levels including the attribute, alternative and global levels. The feedback model is used to solve an engineering project management software selection problem to demonstrate its detailed implementation process, its validity and applicability, and its advantages compared with the GC based approach.     

An intercausal cancellation model for Bayesian-network engineering

When constructing Bayesian networks with domain experts, network engineers often use the noisy-OR model, and causal interaction models more generally, to alleviate the burden of probability elicitation: the use of such a model serves to reduce the number of probabilities to be elicited on the one hand, and on the other hand forestalls experts having to give assessments for probabilities with compound conditions which they feel are hard to envision. Recently, we have shown that ill-considered use of the noisy-OR model specifically can substantially decrease a network's performance, especially in domains in which causal mechanisms include cancellation effects. Motivated by this observation, we designed a new causal interaction model, with the same engineering advantages as the noisy-OR model, to describe such effects. We detail properties of our intercausal cancellation model, and compare it against existing causal interaction models. We further illustrate the application of our model in the real-world domain of pharmacology.     

Semiqualitative modelling of bioprocesses

This paper deals with semiqualitative modelling of bioprocesses with a view to their supervision. An analysis of several approaches for modelling shows the difficulties involved in taking into account in a same framework, quantitative and qualitative knowledge, generally available about a process that we want to control. We propose an original approach, placed in the context of semiqualitative modelling, that is supported by a knowledge model the variables and parameters of which are defined by intervals. For these semiqualitative models, we study their properties in simulation and prediction, and more precisely, their fitting based on experimental data. We show that pertinent predictions in a short time can be obtained, making of these semiqualitative models interesting tools for the development of systems for bioprocess supervision     

Prolegomena to OR Modelling of the Global Environment-Development Problem

The environment is a topical global issue. It concerns the reconciliation of industrial development and environmental protection. Much investigation has been carried out using the numerous scientific modelling tools and techniques available. Some of these modelling tools and techniques belong to the field of Operational Research (OR). This paper proposes an OR approach to address the global environment-development (ED) problem in its broadest possible terms. It sketches out broadly the building blocks of a proposed methodological framework for modelling the ED problem.     

Conceptual modelling and the project process in real simulation projects: a survey of simulation modellers

A survey was used to obtain information on the processes and methods used by simulation experts in real projects. The 102 survey respondents answered questions about their most recent simulation project. This paper presents some of the survey results, focussing mainly on conceptual modelling and the pattern of time allocation to different topics. There are a wide range of findings that include the modellers making changes to the initial conceptual model during subsequent tasks in most of the projects usually by adding complexity, model coding taking on average about twice the time of other topics, and the topics generally occurring in single blocks of time (at the resolution of the survey data collection) but with considerable overlaps. The results give an insight into the way experts approach simulation projects and their problem solving strategies. A potential application is in training novice modellers, particularly in developing ‘craft skills’. The results also provide an empirical basis for further research, especially in conceptual modelling.     

Competitive dynamics and gaming simulation: lessons from a fishing industry simulator

Experiments with a gaming simulator of the fishing industry show that a wide range of firm and industry performance can arise from players’ differing perceptions of the competitive environment. When transferred into the arena of applied strategy development, the results suggest that modellers should give more attention to modelling alternative conceptualizations of strategic intent in the minds of rival business leaders and the firms they create. This new interpretive emphasis should complement the traditional modelling of cross-functional coordination that has been the hallmark of much published work in strategic modelling and simulation. When leaders and firms in the same industry adopt quite different views of the overall system of resources in the industry, it is important to model the heterogeneity of rival firms in order to understand the dynamic performance of the firm and the industry. We propose a modelling approach that captures heterogeneity among the strategic resources that rival firms seek to build and in the operating goals and coordinating processes they use to build them.     

On the drawbacks of large committees

This paper aims at identifying possible drawbacks of large committees. I present a model of a committee of experts with career concerns where each expert may observe an argument about the state of the world but be unsure about the argument’s soundness. Experts may remain silent or decide to be the first to announce an argument. In contrast with the existing literature, I focus on the possibility that experts may discover duplicate information. I show that experts become more reluctant to speak in larger committees as a result. This effect can be sufficiently strong to make small groups of experts optimal.     

Imbalance-oriented SVM methods for financial distress prediction: a comparative study among the new SB-SVM-ensemble method and traditional methods

Previous studies on financial distress prediction (FDP) almost construct FDP models based on a balanced data set, or only use traditional classification methods for FDP modelling based on an imbalanced data set, which often results in an overestimation of an FDP model’s recognition ability for distressed companies. Our study focuses on support vector machine (SVM) methods for FDP based on imbalanced data sets. We propose a new imbalance-oriented SVM method that combines the synthetic minority over-sampling technique (SMOTE) with the Bagging ensemble learning algorithm and uses SVM as the base classifier. It is named as SMOTE-Bagging-based SVM-ensemble (SB-SVM-ensemble), which is theoretically more effective for FDP modelling based on imbalanced data sets with limited number of samples. For comparative study, the traditional SVM method as well as three classical imbalance-oriented SVM methods such as cost-sensitive SVM, SMOTE-SVM, and data-set-partition-based SVM-ensemble are also introduced. We collect an imbalanced data set for FDP from the Chinese publicly traded companies, and carry out 100 experiments to empirically test its effectiveness. The experimental results indicate that the new SB-SVM-ensemble method outperforms the traditional methods and is a useful tool for imbalanced FDP modelling.     

On the modelling of nested risk-neutral stochastic processes with applications in insurance

We propose a modelling framework for risk-neutral stochastic processes nested in a real-world stochastic process. The framework is important for insurers that deal with the valuation of embedded options and in particular at future points in time. We make use of the class of State Space Hidden Markov models for modelling the joint behaviour of the parameters of a risk-neutral model and the dynamics of option market instruments. This modelling concept enables us to perform non-linear estimation, forecasting and robust calibration. The proposed method is applied to the Heston model for which we find highly satisfactory results. We use the estimated Heston model to compute the required capital of an insurance company under Solvency II and we find large differences compared to a basic calibration method.     

Psychological aspects of equation-based modelling

Psychological aspects of equation-based modelling languages like Modelica are under-represented in literature. This does not reflect the growth of the corresponding userbase. In this paper we try to close this gap by tackling the problem from three sides: we conduct expert interviews, we conduct an experiment based on self-reported timings to analyse the effects of inheritance on understandability, and we conduct an online experiment to analyse the effects of model representations on the performance at modelling tasks. The expert interviews suggest that experienced modelling experts tend to develop their models from the top-down, while novices do the opposite. Results from the second experiment indicate that the effect of inheritance on the time to understand a model is both significant and large. The results of the last experiment imply that graphical representations outperform block-diagrams for several metrics. These results open a broad research field on the theory of good modelling practice.