Plenary Lecture

Scenario and Risk Management Simulation for Supporting Strategical Operational Management in Process and Manufacturing Industry

Professor Roberto Revetria
Docente di Impianti Industriali e Meccanici
Italy

Abstract: Modeling and simulation is a powerful tool able to address complex decision making process: it allows the responsible to infer the effect of a decision well in advance and to avoid excessive risky decisions. In literature there several Simulation Methodologies are described, in particular Discrete Event Simulation (DES) is the traditional paradigm used for model manufacturing processes, in such view the problem is describe in term of Entities (nouns) , Resources and Activities (verbs). Several application examples of such vision can be identified in the Steel industry, we have models of steel melting process, of steel rolling mills and of steel treatments. Generally speaking these examples rely on a very hi-fidelity hi-detailed model of the production process that requires a very deep knowledge of the process in order to be properly applied.
There are, however, another class of modeling methodologies that are more suitable for strategic planning evaluation rather than for operative/tactical simulations; in such approaches the experimenter create an approximate model of the systems in order to investigate, at a very high level, what could happened to the system in specific cases (scenario simulation). The focus of the simulation is then the robustness of a solution rather that its performances, in literature these example have been extensively presented and discussed: they are known under various names: Monte-Carlo Simulations, Risk Analysis Tools, Strategic Simulations, Worksheet Simulations and many others.
The basic concept underlying this methodology is founded on the creation and evaluation of several scenario according to specific multi-disciplinary criteria: a long term investment plan in the steel industry could be tested against the occurrence of several events (market growth, economy recession, new players appearances, etc.). The simulator creates and evaluate thousands of different “stories” each one carrying one or more occurrences and test the response of the stimulated system to these “events”. In the following bulleted list is possible to identify possible application of such methodology:

• Product mix decisions where demand and resource requirements for each product are uncertain

• Optimal inventory ordering decisions where product demand is uncertain

• Capacity planning and optimal configuration of plant capacities when faced with uncertain demand for a new product

• Capital budgeting and project selection when resources required for available projects are uncertain

• Optimal cash management policy where cash inflows are uncertain

• Capacity planning for utilities where power usage and power prices are uncertain

• Timing market entry decisions where market growth is uncertain, including the possibility of reduced market share due to late entry and danger of limited market growth

• Making test-marketing decisions that maximize product profitability in the face of uncertain demand

Anandalingam (1987) proposed the process model of the Indian iron and steel industry with particular focus on energy use and conservation possibilities. Uncertainty problems in the industry where analyzed by using probabilistic Monte-Carlo simulations, where model parameters where modeled to be stochastic and distributed normally.
There is another type of simulation that build on top of the Theory of Systems developed by MIT Professor Jay W. Forrester at the Sloan School of Management, this approach is known as System Dynamics and could be used to create Decisions Cockpits able to dynamically evaluate the effect of a strategical decision. Based on the case of a two-echelon steel industry supply chain, Hafeez et al. (1996) demonstrate the application of Systems Dynamics to supply chains and describe an integrated system dynamics framework, with the aim of giving an example to good total systems design. The modeling exercise deals with the design of a supply chain with respect to moving more rapidly towards a minimum reasonable inventory, whereby the chain exhibits capacity constraints, breakdowns and material supply lead-time bottlenecks.
Over the past twenty-five years there has been a significant reduction in UK iron and steel making capacity. This has happened primarily as a result of a deliberate strategy which places emphasis on maintaining or improving financial performance. Dangerfield and Roberts (2000) described the use of a System Dynamics simulation model able to identify an unanticipated but possible scenario of the consequences of this strategy. The model considered both short and long term effects and represented and aid to learning in the face of the complexity which characterizes manufacturing operations.
In Author’s research group several experiences have been achieved by applying scenario and risk management simulation to real industrial cases, in every application great attention was posed to the integration phase in order to create a Decision Support System (DSS) constantly updated with the real data collected by the Enterprise Data Information System.

• Development of the applicative method and use both Monte-Carlo and Artificial Intelligence techniques (Fuzzy Logic, Neural Networks, Genetic Algorithms) for the solution of off-line and on-line problems of stock management (1998-AMT-SGS Project, Azienda Mobilita e Trasporti) and Production Scheduling (2000-FRINE Project, Marconi Telecommunications).

• Development of Monte-Carlo models of analysis for complex Power Plant maintenance with the PUMA 2001-Project for Ultimate Maintenance in ANSALDO and 2002-TARAS (Finmeccanica Group) Projects.

• Development of an urban network optimization model for the waste collection and treatment, by considering also the extension and application to the differentiated collection with the development of algorithms aiming to the construction of the optimal path and the collection vehicle fleet dimensioning (2006-PING Project).

• Development of a System Dynamics Model for supporting complex decision making in Pharmaceutical Industry (2007-IB Informatica)

• Development of a Monte-Carlo model for supporting the Risk Identification and Mitigation in dangerous goods logistics and transportation (2006-Provincia di Savona Project)

• Development of a simulation meta-model for supporting the investments stream allocation in manufacturing industry (2005-Whirlpool Europe Project)

The possibilities offered to Process and Manufacturing Industry by Monte-Carlo and System Dynamics simulation are many, in particular is possible to create a model for supporting the evaluation of an investments plan by identifying the more promising investment streams within a constrain of a limited budget, in this way the model will identify the most robust strategy able to maximize profits. The implementation of a Decision Cockpit is another possibility offered by the system Dynamics methodology, the simulator will be connected to the company ERP (i.e. SAP R/3) that will keep “warm” the mod el with the real life system allowing complex what if scenarios evaluation through the use of a very user friendly graphical user interface. The user will have “handles”, “sliders” and “joysticks” to control the model and to interact with it obtaining the real-time reaction in a simple widget panel.
There are other possibilities that can be explored based upon a specific needs since scenario and risk modeling and simulation remains one of the most promising methodology in the field of the Strategical Operation Management.

Brief Biography of the Speaker:
In 1991 he takes the Commercial School leaving diploma of Accountant and Estimator and then holds the role of administrator for “RPM snc di Revetria P&R Costruzioni Impianti Frigoriferi Industriali” Company. In 1997 he participates to ICAMES with the WOLVES project winning the Prize for the best algorithm about the Prisoner Dilemma “Bush-8”.
He gets the degree with honours in 1998 and wins the 4° Competition for Italian Navy Reserve Officer Cadet; in July 1999 he is appointed Acting Sublieutenant in the Naval Army Corps. From July 1998 to July 1999 he is attached to the Control and Testing Service Head Office (Capo Servizio Controllo e Collaudi) of the Military Maritime Arsenal of Taranto where he holds the role of Responsible for the on land bodies Control and Testing. During this period he develops a strong experience not only in management and testing of the Technical Plants (Desalination Plants, Frigorific Cells, Sewage Drainage Plants) and Civil Works (Building up of 3 lighthouses, re-engineering of the Artillery Maintenance Department (Settore Manutenzione Artiglierie), but also in testing of on Board Scaffolds and in re-organisation of the Control and Testing Service of the on Land Bodies, coordinating 15 persons. During 1998 he wins the Scholarship for the XIV Cycle Doctoral Program at the University of Parma - Faculty of Engineering which ends with success in October 2001 by getting the title of PHD. On December 20, 2000 he passes the exam for the Level D Project Manager IPMA Certificate. On November 2001 he wins the Competition for the role of Researcher at the University of Genoa where he takes service at the Department of Production Engineering. The personal experience achieved in the construction of computer-aided analysis models and tools brought him to develop complex software using multipurpose (C/C++, Java, Tcl/Tk, Python, Php, Fortran, Pascal, VBA) and dedicated (Automod, Simul8, ESL, GPSS/H, Arena, Witness) languages. Up today he has been working in the development of the mentioned software above all in the Windows, Linux and Mac environments by acquiring experience on the main applications and office automation packs. He makes use and develops models in virtual reality for the Virtual Design by exploiting also Web-Oriented graphic environments (Atmosphere, Java 3D, Blender, Vega Prime, Multigen-Creator). On March 2004 he wins the Competition for the role of Associated Professor at the University of Parma. He is, as well, Associate Director of the McLeod Institute of Simulation Science (MISS) - San Diego CA, USA.