WSC 2003

WSC 2003 Final Abstracts

Business Process Modeling/Reengineering Track

Monday 10:30:00 AM 12:00:00 PM
The Role of Special Agents in Today’s World

Chair: Craig Robertson (Accenture)

Agent-Based Modeling and Simulation of Store Performance for Personalized Pricing
Cem Baydar (Accenture Tech Labs)

In this paper, a simulation-based approach of optimizing a grocery store’s performance is discussed. Currently, most of the grocery stores provide special discounts to their customers under different loyalty card programs. We believe that a more determined approach such as personalized pricing could enable retailers to optimize their store performance. The objective of this paper is to determine the feasibility of personalized pricing to optimize store performance and compare it with the traditional product-centered approach. Each customer is modeled as an agent and his/her shopping behavior is obtained from transaction data using factors such as customer’s product consumption rate, brand loyalty and price sensitivity. Then, the overall shopping behavior is simulated and the store performance is optimized. The results showed that personalized pricing outperforms the traditional product-centered approach significantly. It is expected that successful implementation of this work will impact grocery retail significantly by increasing the customer satisfaction and profits.

Combining Agent-Based Supply Net Simulation and Constraint Technology for Highly Efficient Simulation of Supply Networks Using APS Systems
Hartwig Baumgaertel and Ulrich John (DaimlerChrysler AG)

This paper introduces an approach for highly efficient simulation of supply networks in which several nodes use advanced planning and scheduling (APS) systems. APS systems increasingly become part of bigger companies IT landscape. Today, APS systems communicate only with the ERP system they sit on top of. If APS systems would be able to communicate with each other, qualitatively new processes for planning collaboration in supply nets could emerge. Simulation is accepted to be an useful approach for support of business process design. Design of planning processes which should exploit APS systems requires simulation systems with integrated APS functionality. We augmented an existing, agent-based simulation system by an APS component which we developed based on finite domain (FD) constraint technology. In this paper, we present our simulation system with special focus on the APS component, and results of a simulation experiment which was used for the proof of concept.

Simulating Agent Intelligence as Local Network Dynamics and Emergent Organizational Outcomes
James K. Hazy and Brian F. Tivnan (The George Washington University)

We build upon our previous work (Hazy and Tivnan 2003) to represent organizations as a network of agents, tasks, resources and knowledge (Krackhardt and Carley 1998) to explore the emergent effects of agent interactions on organizational outcomes. To do this, we define agents in the context of their position in the network, describe the agent’s symbolic representation of its position in the network, and develop a probabilistic function associated with each agent that acts locally to change the network. We conclude with a brief overview of our research in this area to date and the usefulness of this network representation.

Monday 1:30:00 PM 3:00:00 PM
The Process of Process Reengineering

Chair: Kishore Swaminathan (Accenture)

Visualization of Probabilistic Business Models
Lev Virine and Lisa Rapley (Schlumberger Ltd.)

One of the main challenges in the modeling of business problems is to provide the modeler and the user with meaningful visual tools. The business model is usually presented by different types of flow charts and diagrams. If the modeling process is simplified in how it is represented to the user, it improves understanding, as well as, helps to interpret the result of the analysis. This paper discusses a proposed methodology for business modeling and how this process can be applied to real world problems. The formal iterative modeling process includes a Probabilistic Model Description, Domain Model Diagram, and diagrams to define model’s calculation logic, sensitivity analysis tools, decision trees, and other tools. The paper also discusses benefits of the unification of specification for the visualization tools. The described methodology is used in decision and risk analysis application Decision Tool Kit.

Simulation for Business Processes and Information Systems Design
Ray J. Paul and Alan Serrano (Brunel University)

Business Process (BP) literature promotes the value of business processes as essential gearwheels that help organizations to reach their goals. Similarly, many process design approaches claim that Information Technology (IT) is a major enabler of business process, a view also shared by the Information Systems (IS) community. Despite this, BP and IS approaches do not provide clear guidance on how to assess the benefits that a given IS design may bring to the BP prior the IS implementation. Nor is clear indication of which modeling techniques could be used to assess such relationship. This paper uses the insights gained during a UK funded research project, namely ASSESS-IT, that aimed to depict the dynamic relationships between BP and IT to propose an alternative framework to develop BP simulation models that depict the dynamic behavior of the relationships between BP and IS.

Integration of Discrete Event Simulation Models with Framework-Based Business Applications
Peter Lendermann, Nirupam Julka, Lai Peng Chan, and Boon Ping Gan (Singapore Institute of Mfg. Technology)

Simulation models and business application software as they are used for decision support in enterprise management are both representations of an enterprise’s actual operations. This paper describes a unified simulation and application framework where it is possible to represent the entire performance process along a supply chain in a unified business model, improve its performance with discrete event simulation technology, and then generate and implement the corresponding business application software from the same unified model, based on a so-called framework-based application technology which allows implementation of changes derived from simulation analysis with minimal effort and time. This enables a company to optimise not only operational processes such as shopfloor or warehouse operations but also business processes such as planning, order management and scheduling through simulation.

Monday 3:30:00 PM 5:00:00 PM
Customer Relations Management: Call Center Operations

Chair: Pierre L'Ecuyer (University of Montreal)

Modelling and Simulation of a Telephone Call Center
Juta Pichitlamken, Alexandre Deslauriers, Pierre L'Ecuyer, and Athanassios N. Avramidis (Université de Montréal)

We consider a system with two types of traffic and two types of agents. Outbound calls are served only by blend agents, whereas inbound calls can be served by either inbound-only or blend agents. Our objective is to allocate a number of agents such that some service requirement is satisfied. We have taken two approaches in analyzing this staffing problem: We developed a simulation model of the call center, which allows us to do a what-if analysis, as well as continuous-time Markov chain (CTMC) queueing models, which provide approximations of system performance measures. We describe the simulation model in this paper.

Routing Heuristics for Multi-Skill Call Centers
Ger Koole and Auke Pot (Vrije Universiteit) and Jérôme Talim (Carleton University)

We give an approximation method for analyzing the performance of call centers with skill-based routing, for both blocking and delay systems. We use this method to determine optimal skill sets for call center employees.

Fluid Approximations for a Priority Call Center with Time-Varying Arrivals
Ahmad D. Ridley and Michael C. Fu (University of Maryland, College Park) and William A. Massey (Princeton University)

We model a call center as a preemptive-resume priority queue with time-varying arrival rates and two priority classes of customers. The low priority customers have a dynamic priority where they become high priority if their waiting-time exceeds a given service-level time. The performance of the call center is measured by the mean number in system for the two customer classes. A fluid approximation is proposed to estimate the mean number in system for each class. The quality of the approximation is tested by comparing it with a stochastic simulation model of the system. Finally, using the fluid approximations, we discuss how to compute the mean number in system for each class and estimate the overall staffing level, or number of agents.

Tuesday 8:30:00 AM 10:00:00 AM
Customer Relations Management: Service Operations

Chair: Cem Baydar (Accenture)

Using Simulation to Approximate Subgradients of Convex Performance Measures in Service Systems
Július Atlason and Marina A. Epelman (University of Michigan) and Shane G. Henderson (Cornell University)

We study the problem of approximating a subgradient of a convex (or concave) discrete function that is evaluated via simulation. This problem arises, for instance, in optimization problems such as finding the minimal cost staff schedule in a call center subject to a service level constraint. There, subgradient information can be used to significantly reduce the search space. The problem of approximating subgradients is closely related to the one of approximating gradients and we suggest and compare how three existing methods for computing gradients via simulation, i.e., finite differences, the likelihood ratio method and infinitesimal perturbation analysis, can be applied to approximate subgradients when the variables are discrete. We provide a computational study to highlight the properties of each approach.

Simulation's Role in Baggage Screening at the Airports: A Case Study
Suna Hafizogullari, Gloria Bender, and Cenk Tunasar (TransSolutions)

The Aviation and Transportation Security Act passed by Congress in November, 2001 required the nation’s airports to perform 100% checked baggage screening by December 31, 2002. To determine the impact of this requirement on its operations, Lambert St. Louis International Airport (STL) requested TransSolutions to evaluate the equipment and facility requirements to meet 100% checked baggage screening for all airlines serving STL. Discrete event simulation models were developed to evaluate passenger service levels for each alternative option considered, relative to the airport performance metric that 95% of all passengers in the peak hour would wait no longer than additional 10 minutes for baggage screening. Various protocols with different machine requirements were tested, and the “Drop-and-Go” option was chosen as the most viable alternative. This paper discusses how simulation was used to help the airport’s decision making process.

Human Fatigue Risk Simulations in 24/7 Operations
Rainer Guttkuhn, Udo Trutschel, Anneke Heitmann, Acacia Aguirre, and Martin Moore-Ede (Circadian Technologies, Inc.)

A Circadian Alertness Simulator (CAS) is presented as an interactive tool for assessing sleep behavior and fatigue risk in the 24/7 operations. The simulation model uses as input sleep-wake data and information about individual sleep characteristics (short vs. long sleeper, morning type vs. evening type, napper vs. non-napper). The validation of the CAS model was based on a figure of merit function reflecting the model’s ability to minimize the difference between reported and predicted sleep data. The purpose of the alertness model is the assessment of work schedules in terms of fatigue risk.

Tuesday 10:30:00 AM 12:00:00 PM
Process Improvement

Chair: Eleazer Martin (BestSys)

Optimization of a Telecommunications Billing System
Mark Schouwenaar and Eleazer Martin (BestSys)

To remain competitive in a turbulent and rapidly evolving market, telecommunication companies have found it necessary to invest large sums of money on the latest technologies and IT infrastructure. These investments have been a serious drain on the financial resources of these companies who are now seeking ways to pare costs and regain their financial footing. This new reality is increasingly forcing companies to focus on improving processes in order to increase profitability. Process simulation is proving to be a useful tool in helping these companies attain higher levels of efficiency in business critical processes by revealing inefficiencies and redesigning processes. This paper seeks to illustrate the method used to obtain substantial savings in the billing system of Telenor (Norway’s largest telecom company), through the use of simulation.

One Size Fits All? Segmenting Customer base for Maximum Returns
Roar Grønhaug and Eleazer Martin (BestSys) and Per-Åge Bæra (Telenor)

Large corporations can achieve significant cost savings by developing and employing a sophisticated and continuously updated, billing and credit policy. Days of sale outstanding (DSO) is a major cost driver for corporations with large revenues, as this leads to an increased risk of default, increased dunning and collection costs, a non-optimal billing procedure with attendant costs and perhaps most importantly, an increase in the order-to-cash cycle time and the significant increase in hidden costs this implies. Segmentation of the customer base according to behavior and risk combined with the design of bespoke billing and credit policies suited to the behavior and risk associated with each segment, can lead to a significant decrease in the costs mentioned above. This paper illustrates the work done at Norway’s largest telecommunication operator, Telenor, to address these issues using the continuous simulation methodology as well as other econometric tools.

Simulations on .Net Using Highpoint's HighMAST™ Simulation Toolkit
Peter C. Bosch (Highpoint Software Systems, LLC)

This paper describes the philosophy, architectures and key features of a new .Net-based simulation object model and toolkit called HighMAST™ (Highpoint Modeling and Simulation Toolkit). HighMAST™ is a set of class libraries built on top of Microsoft’s .Net platform. It was built to take advantage of the object-oriented flavor and extensive integration plumbing ingrained in the .Net framework. It supports “active entity”, “block based”, “workflow oriented” and several other types of simulation architectures in both the discrete-time and continuous domains. And it enables developers to approach their simulation frameworks or applications in a wide range of languages including such inexpensive and available languages as C# and VB.Net.

[ Return to Top | Return to Program ]