McLeod Institute of Simulation Sciences
Academic Certificate

PhD Degree given by the

University of Calabria

To:

Francesco Longo

Supervisor:

Enrico Papoff

No:

3

Thesis Title

M&S Applied to Design and Management of Production and Logistics Systems

Summary
The thesis presents the results of the research activities developed during the three-years Doctorate courses at Department of Mechanic of University of Calabria, Industrial Engineering Sector. The Doctorate is concerned with Modeling & Simulation (discrete event simulation) applied to manufacturing systems and logistic systems.
Even though these research topics, as clearly summarized in chapter 1 (refer to the state of art overview of the different research areas), cannot be considered “young” in terms of research works developed and relative publications and considering that before this Doctorate no other works had been developed in this specific area at Industrial Engineering Sector of University of Calabria, it has resulted pretty interesting to focalize on such subjects aiming to reach two fundamental objectives summarized as follows.
In other words the research activities developed during the Doctorate raises two important issues regarding the possibility of using Modeling & Simulation as methodology to obtain new scientific results and the opportunity of using Model & Simulation as innovative tool or innovative approach to face classical problems. Looking to the thesis objectives it’s simple to understand that the first objective is complementary to the second and vice versa; this allows to become knowledgeable for what concern this subject as much as possible.
The first Doctorate logic step was the selection of the specific research areas using the academic and scientific definition of Industrial Engineering Sector, reported as follows.
“The sector studies the methodologies and the general criteria related to production systems planning, designing and management. The sector groups the following scientific areas: production systems analysis and design, including business plans, plant location methodologies; plant general services analysis and design including technical and economic optimization methods; production and processes technology analysis and design; workplaces and production systems ergonomic design; production systems management including quality and maintenance; production systems internal and external logistic, including material handling; production systems automation including economic analysis of integrated and flexible systems and industrial tools for automated controlling process.”
The Industrial Engineering academic definition highlights some important research areas as the production systems management and external logistic. Looking in detail to these research areas we decided to focalize on the following topics.
It’s important to underline that short period production planning and plant lay-out analysis and optimization are strictly related and considered as two of the most important factors to improve production system efficiency. The same consideration can be made for what concern the focuses on external logistics.
In addition, as we will explain in detail in the chapters of the thesis, it has been found out that these topics are suitable research areas to obtain pretty interesting scientific results and test innovative ways and approaches to use Modeling & Simulation.
The Doctorate activities have been subdivided in three parts: training activities, state of art overview and research activities.

Training activities
Training activities aimed to learn all the steps of a simulation study and the software tools to develop a simulation model. Detailed studies have been made concerning simulation modeling principles, input data analysis, Verification Validation & Accreditation (VV&A), simulation runs planning using Design Of Experiments (DOE), simulation output analysis using Analysis Of Variance (ANOVA).
Several and different software tools have been learned and used to create the simulation models. eM-Plant by Tecnomatix Technology (a discrete event simulation package), Anylogic by XJ Technologies (a java-based simulation package), Creator by Multigen (a 3D tool for real-time simulation), Vega Prime (a 3D virtual environment for II
simulation), Minitab by Minitab Inc. (a specific tool for statistical analysis), C++, Java, Visual Basic, Simple ++ programming languages (for developing simulation model directly by code).
Training activities have been made at University of Calabria, Industrial Engineering Sector studying the simulation modeling principles and input data analysis learning at the same time eM-Plant, Simple++, Visual Basic and Minitab; at University of Genoa (Savona campus, Industrial Engineering Sector) in collaboration with MISS (McLeod Institute of Simulation Science) and Liophant Simulation Club for what concern VV&A, DOE and ANOVA and learning Creator,
Vega, C++, at Rutgers University (the State University of New Jersey, Industrial Engineering Department) in collaboration with DIMACS Center (Discrete Applied Mathematics Center) for what concern simulation modeling principles, input data analysis, ANOVA and learning Anylogic, JAVA and Minitab.
The results of the training activities can be observed in each chapter of the thesis, where, all the studies as well as all the software and programming languages have been applied in the research activities.
State of art overview In order to start the research activities an overview of the state of art in the specific research areas has been made reading several books, international journal papers and conferences papers. The results of this phase are summarized in chapter 1.

Research activities
The research activities have been subdivided in case studies concerning the two specific research areas before mentioned. In particular chapter 2 and chapter 3 deal with production systems management. Chapter 2 focalizes on short period production planning presenting two cases study. Both cases study consider the jobs scheduling problem in manufacturing systems. This type of problem is in general faced neglecting the stochastic nature of production systems, in other words proposing a deterministic approach and maintaining several restrictive assumptions (see chapter 2, paragraph 2.1), trying to avoid highly complex and computationally intractable problems. The objective of our research is to demonstrate how all the restrictive assumptions can be deleted using Modeling & Simulation to recreate systems complexity. The fundamental idea is to make good assumptions to have a good model and not to have an easy solvable problem from a theoretical point of view. Following this way we don’t aim to obtain results with general validity (appropriate for each production system, this could be almost impossible due to the huge differences between real manufacturing systems) but to obtain result that can be easily transferred to the real systems. In addition chapter 2 (the second case study) describes an “alternative” way to use a commercial discrete event simulation package by-passing the classic object oriented architecture (during the modeling phase), writing code to develop the simulator and using tables to manage model information.
Chapter 3 proposes two cases study regarding plant lay-out analysis and material handling systems. It has been found out the simulation potentialities as cognitive tool (for manufacturing systems still in the design phase) and, skipping the traditional methods for material handling systems design, it has been used (for this purpose) a genetic algorithms-based approach. Besides, the second case study presents a plant lay-out optimization of a real manufacturing system. Due to the complexity of the problems that characterized the analysis and optimization of plant lay-out, considering the high stochastic nature of the main variables (e.g. the process and set-up times, market demand variability and so on) as well as the company requirements to have feasible solutions, it’s extremely clear the inadequacy of traditional methodologies for these type of studies and analysis. We will show that simulation plays a critical role as problem solving tool that can be used to test several and different scenarios and perform what-if analysis, taking into account the complexity of the systems and proposing optimal solutions (also in this case a genetic algorithms-based approach is proposed).
Chapter 4 and chapter 5 are concerned with external logistic. Chapter 4 presents three cases study focalizing on some priority aspects of the logistic side of the supply chain as the design parameters in the transportation systems, proposing an ANOVA-based selection policy for adding new supply chain customers and presenting a completely innovative approach to face the classic problems related to the inventory/store management of a supply chain facility (in this case no other research works have been found proposing something similar). Throughout the chapter we experiment again the “alternative” way to use a commercial discrete event simulation package (in order to obtain flexible simulator) and we propose how to integrate different software and C++ programming to create a 3D simulator.
Finally chapter 5 focuses on security side in logistic systems and supply chain as natural consequence of September the 11th, presenting one case study. Obviously this research area is still “young” (most of the research works have been developed after 2001) and we have found out that Modeling & Simulation is a powerful tool to face this type of problems. In particular the research presents a java-based simulator to model the operations of a terminal container and analyze security procedures for containers inspection and their impact on container terminal performances.
The research activities has been developed at University of Calabria, Industrial Engineering Sector (chapter 2, cases study 1, 2; chapter 3, cases study 1, 2; chapter 4, case study 2) at Genoa University, Savona Campus, Industrial Engineering Sector (chapter 4, cases study 1, 3) at Rutgers University, the State University of New Jersey, Industrial Engineering Department (chapter 5, case study 1).

Back to MISS Certificates