Roberto Mosca

Agent directed HLA simulation for complex supply chain modeling

The authors outline a new modeling methodology designed to simulate a complex logistics network and to ensure interoperability through application of Intelligent Agent High-Level Architecture for distributed supply chain management.

Risk analysis in harbor environments using simulation

The paper proposes the development of an integrated interactive environment for risk analysis applied to harbor and maritime environments. This approach is very effective for properly designing the harbor and maritime infrastructure in terms of resources, structures and services for facing possible emergencies. The authors describe the general architecture of this tool and the integration with a specific simulation module devoted to an oil spill study and analysis, including experimental results and statistical validation.

Monte Carlo Simulation Models Evolving in Replicated Runs: A Methodology to Choose the Optimal Experimental Sample Size

The idea of a methodology capable of determining in a precise and practical way the optimal sample size came from studying Monte Carlo simulation models concerning financial problems, risk analysis, and supply chain forecasting. In these cases the number of extractions from the frequency distributions characterizing the model is inadequate or limited to just one, so it is necessary to replicate simulation runs many times in order to obtain a complete statistical description of the model variables. Generally, as shown in the literature, the sample size is fixed by the experimenter based on empirical assumptions without considering the impact on result accuracy in terms of tolerance interval. In this paper, the authors propose a methodology by means of which it is possible to graphically highlight the evolution of experimental error variance as a function of the sample size. Therefore, the experimenter can choose the best ratio between the experimental cost and the expected results.

Simulation as support for production planning in small and medium enterprise: a case study

The proposed application is related to an Italian small factory that produces, assembles, and sells mechanical components for awnings. In such factories BPR (business process engineering) usually takes place without the support of modeling & simulation although such methodologies have proved to be very effective and helpful. Main reasons for that have to be investigated in the high costs usually associated with a simulation study, especially for data collection, model building and model validation. In order to avoid this problem a general-purpose simulation framework was designed enabling self-build according to production process information stored in a relational database. Moreover, the simulation model was used in conjunction with a statistical analysis tool in order to build the relationship among selected parameters and the proposed objective function by mean of response surface methodology and 2nd order regressions meta-models. Authors applied the proposed schema to several industrial applications obtaining interesting results.

Simulation of Dock Management and Planning in a Port Terminal

Abstract Dock and yard space management is one of the main problems in a port container terminal. In this sector, modelling can provide interesting solutions to the problem of verifying current planning. However, it is particularly important to set up this work as the first link in a decision-support chain that not only evaluates the consequences of such planning, but also becomes the active smart proposal developer. This paper presents the first step in this direction.

Evaluation And Analysis By Simulation Of A Production Line Model Built With Back-Propagation Neural Networks

Once a plant simulator had been built, the next objective was to create a simple-to-use model of a system that could provide immediate responses. Neural networks are ideal tools to achieve this objective, and this article presents the results of an experimental study aimed at identifying valid reference criteria for selecting the network architecture that is most suitable for the particular situation examined.

Simulation of Hazardous Material Fallout for Emergency Management During Accidents

This article proposes a model to evaluate hazardous material fallout. The problem is particularly pertinent since the transport, storage and production of this type of product often occur close to inhabited areas and, in case of accidents, it is important to correctly estimate the lines of action and risk areas. This model can also be used to evaluate the environ mental risk related to specific hazardous installations and therefore can be used to improve industrial planning.

Simulation-based VV&A methodology for HLA federations: an example from the Aerospace industry

This paper establishes a methodology for the validation of a whole federation of simulation models which represent the reality of the supply chain in the context of the aerospace industry. There are many possible ways to address this problem, among these, the simplest way appears to be the analysis of the real-life situation, based on historical data. The use of historical data is not always a viable approach, due to the lack of a sufficiently extensive pool of reliable data. Therefore a simulation-based VV&A approach is often required before the operative use of models. Historical data is reproduced through a long series of runs of a simulation meta-model. The major issue associated to the validation of a federation of simulation models is that each simulation model needs to be validated as stand-alone and at the same time it needs to be validated as a member of the whole federation, or of a portion of it.

Risk analysis in the evaluation of plant investments: the contribution of a non-deterministic approach

This paper discusses the application of risk analysis techniques to a feasibility study for a chemical additives production plant in Saudi Arabia. The ordinary estimation techniques, which indicate utilization of resources, future cost trends, and operating profit forecasts, are supported by a risk analysis model that is used to evaluate the variability of analyzed targets based on the definition of different scenarios. Project investors can use this tool during the decisional phase to minimize the risks related to project results.

AI-based optimization for fleet management in maritime logistics

This paper outlines the features of an automated decision support system (DSS) developed to optimize the logistics of maritime transportation for a large chemical company. The paper focuses on the design and implementation of an optimization module to complement a DSS architecture including dynamic databases, decision heuristics, and dynamic process simulation, for the systematic generation of cost-effective fleet configurations capable of meeting the companys production requirements.