Conceptual Modeling (CM) is a fundamental step in a simulation project. Nevertheless, it is only recently that structured approaches towards the definition and formulation of conceptual models have gained importance in the Discrete Event Simulation (DES) community. As a consequence, frameworks and guidelines for applying CM to DES have emerged and discussion of CM for DES is increasing. However, both the organization of model-components and the identification of behavior and system control from standard CM approaches have shortcomings that limit CM’s applicability to DES. Therefore, we discuss the different aspects of previous CM frameworks and identify their limitations. Further, we present the Hierarchical Control Conceptual Modeling framework that pays more attention to the identification of a models’ system behavior, control policies and dispatching routines and their structured representation within a conceptual model. The framework guides the user step-by-step through the modeling process and is illustrated by a worked example.
Planning of operations, such as routing of vehicles, is often performed repetitively in rea-world settings, either by humans or algorithms solving mathematical problems. While humans build experience over multiple executions of such planning tasks and are able to recognize common patterns in different problem instances, classical optimization algorithms solve every instance independently. Machine learning (ML) can be seen as a computational counterpart to the human ability to recognize patterns based on experience. We consider variants of the classical Vehicle Routing Problem with Time Windows and Capacitated Vehicle Routing Problem, which are based on the assumption that problem instances follow specific common patterns. For this problem, we propose a ML-based branch and price framework which explicitly utilizes those patterns. In this context, the ML models are used in two ways: (a) to predict the value of binary decision variables in the optimal solution and (b) to predict branching scores for fractional variables based on full strong branching. The prediction of decision variables is then integrated in a node selection policy, while a predicted branching score is used within a variable selection policy. These ML-based approaches for node and variable selection are integrated in a reliability-based branching algorithm that assesses their quality and allows for replacing ML approaches by other (classical) better performing approaches at the level of specific variables in each specific instance. Computational results show that our algorithms outperform benchmark branching strategies. Further, we demonstrate that our approach is robust with respect to small changes in instance sizes.
Summary
Many health delivery services have required performance targets. Typically, these targets are presented as percentiles of patients to be seen within specified timeframes. These targets present hospital administrators with a resourcing problem complicated by conflicting objectives: How to minimize costs while maximizing throughput to achieve the performance targets? In this paper, we describe the use of a simulation model to evaluate the effect of changes to staff levels in a cytology department, investigating the trade‐off between staff levels and turnaround times in light of performance targets specified by government.
Standard practice for determining staffing levels in a cytology department uses average workload estimates and does not take into account target performance measures, task variability, and the interruptive nature of the workload of pathologists. We develop a simulation model for pathologist workload within a cytology department in New Zealand. We describe the model construction process that follows the hierarchical control conceptual modeling (HCCM) framework. We use the resulting simulation model to examine the trade‐offs between staffing levels (and associated rosters) and task turnaround time.
The results indicate that consideration of variation in task arrivals is important when considering the effect of staffing levels on turnaround time. Furthermore, as the cytology department is required to meet performance targets that involve maximum service times for a percentile of patients, such an approach is necessary in order to estimate the performance level of a staffing roster.
The classical world-views of discrete event simulation (DES) are event scheduling, activity scanning and process interaction. A fourth approach, the three-phase method, extends activity scanning and is often regarded as another world-view. These world-views provide the theoretical framework for applying DES in practice. However, in health care simulation, practitioners often face modeling challenges where the concepts and methodologies described by these world-views are not able to reflect either the dynamics or the entity flow of the system being modelled. This leads to individualized approaches and solutions that do not build on a unified and standardized theoretical basis. In this paper we present an extension to the activity scanning world-view based on the needs of the health care sector that uses hierarchical control structures as a more general, flexible and powerful tool to define health care DES models. To demonstrate the strength and potential of the approach two ongoing simulation studies are briefly outlined and the benefits of using the new world-view for these models is discussed.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.