Thomas D. Clark

The dynamic structure of management support systems: theory development, research focus, and direction

Using a systems perspective, a conceptual model is developed that encompasses a broad class of systems whose fundamental purpose is the support of managerial actions and decision making. The term management support systems (MSS) is used to label this broad class. This model is based on an extensive review of the relevant literature and available research. The result provides an integrated systems model of the phenomena involved and points to gaps in the research that arise largely from the attempts to examine various classes of MSS as separate entities. The research presented here is based on the premise that there are fundamental core consistencies or similarities among various types of systems that have evolved in the past several decades to support decision making. It presents a conceptual, theoretical model drawn from findings about various types of support systems described in the literature such as decision support systems (DSS), executive information systems (EIS), knowledge management systems (KMS), and business intelligence (BI). Pragmatic insights are provided by the conceptual model and recommendations for future research are discussed.

Using system dynamics to anticipate the organizational impacts of outsourcing

The outsourcing of information systems functions is a phenomenon that is now considered an essentially permanent addition to the cache of options available to the modern manager. Yet, while the potential costs and benefits of outsourcing are widely debated, a thorough understanding of the phenomenon remains elusive. The outsourcing decision is highly situational and generalizations regarding its relative merits are difficult to construct. Because of its unique ability to capture the inherent complexity of the outsourcing decision, system dynamics is used to construct a computer-based model of that decision process in a typical information systems organization. The effect of two management policies and two exogenous variables is explored over 80 quarters (20 years). Findings indicate that while outsourcing offers potential cost savings, it is accompanied by a decreased ability to respond to unexpected changes in the competitive market place. Copyright

A simulation analysis of demand and fleet size effects on taxicab service rates

The research presented in this paper was directed at understanding the dynamic interaction between demand, service rate, and policy alternatives in a typical urban taxi system. The general nature of such systems is discussed and a SLAM II simulation model of the system introduced. A series of experiments with the model indicate that customer waiting time is relatively insensitive to changes in demand but highly sensitive to changes in the number of taxi cabs.

Modeling short- and long-term dynamics in the commercialization of technical advances in IT producing industries

Manufacturing industries in the IT sector, which are characterized by very low inertia, rapid technological change, and swift technological obsolescence, are a vivid example of how the rapid and effective commercialization of technical advances is critical to the success of high-technology industries. Radical or breakthrough technologies have both short- and long-term impacts. This paper argues that classical technology diffusion modeling approaches fail to give a fully dynamic picture of technology adoption in an industry, and do not adequately capture dynamic transients caused by short-term trends and the effect of individual technical changes. It then demonstrates the usefulness of the system dynamics approach in modeling the long-term system behavior of the commercialization process arising from the dynamic transients caused by short-term trends and rare events in IT producing industries. Capturing such dynamic transients is critically important because the commercialization process is ultimately shaped by individual decision-makers making the strategic investment decisions within an industry operating in a particular short-term context. Copyright

Taxi management and route control: a systems study and simulation experiment

The structure and behavior of a typical urban taxi system are addressed in this study. It is approached as a complex, first-order feedback system with network queuing structure. A computer simulation model of a characteristic municipal system is proposed and alternative taxi station and routing patterns addressed. An analysis of variance research design is employed to assess model results, and a method for stationing resources and controlling taxi movement is suggested. The simulation model is written in the Simulation Language for Alternative Modeling (SLAM). 1 PROBLEM DESCRIPTION Historically, the taxicab industry has played au important but little recognized role in the provision of public transportation services in urban areas. In many small urban and rural areas, taxicabs have provided the only public transportation services available, and have been an essential service for many community residents without convenient access to an automobile (Kirby, 1981). The form of the taxi service is such that any member of the general public may hire a vehicle and driver to convey the person directly by road from one point to another that is specified by the passenger(s). Taxicabs constitute a form of transportation called psratrsnsit. Kirby et al (1974) define it as “those forms of intraurbau passenger transportation which are available to the public, are distinct from conventional transit (scheduled bus and rail), and can operate over the highway and street system. m It includes those types of transportation in-between the ptivate automobile and conventional mass transit, such as car pools, rental cars, dial-a-ride, jitneys, limousines, and, of cxmrse, taxicabs. Paratransit can be conveniently grouped into three categories: those you hire and drive, those you had or phone, and those for which you make prior arrangements with other tra\’elers. Hail or phone service is often called demand-responsive transportation (DRT) and includes taxicabs and dial-a-ride. In the dial-a-ride problem, customers all a dispatcher or scheduler requesting service. Each customer apecifiti a distinct pickup and delivery point and a desired time for pickup and delivery. Vehicles (usually small busses or vans) are allowed to deviate from their direct route to serve other passengers. This type of system is usually called shared-ride transit (SRT) and is oflen classified as a vehicle routing and scheduling problem. This area has received considerable attention. In a conventional taxicab system, once a vehicle picks up a customer, it proceeds non-stop to the customer’s destination. This is often referred to as exclusive-ride transit (ERT). The operation of an exclusive-ride taxi system is not as easily classified as a scheduling and routing problem. It seems to better be represented as a queuing problem, specifically an M/G/s queuing problem. An MIGIs queuing problem is one in which calls arrive according to a Poisson distribution (M), calls are serviced in accordance with some general service distribution (G), and there are s parallel servers (taxis). Unfortunately, the complexity of the service distribution makes the problem too difficult to address using existing queuing theory unless very restrictive assumptions are made. Even if the service distribution could be derived, exact queuing formulae have not yet been developed for the MIGIs problem. Nonetheless, queuing concepts do provide a useful fi-amework for discussing the problem. A coherent, continuing stream of research that addrwea the opemtion of a taxi fleet apparently does not exist. This is due primarily to the fact that the taxi industry is ao fragmented and private, that it is often t as a viable means of public overlooked by govermnen transit. Thus, little funding for research has been provided, especially when compared with other _tion problems. Only three pieces of reseamh specifically addressing the operation of an exclusive-ride taxi fleet were identi-

A simulation approach to analysis of emergency services and trauma center management

The operation of a hospital trauma center and associated transportation facility is discussed in this paper. A SLAM simulation model of the system is explained and example experiments with it discussed. Results indicate that significant improvement in patient service can be made through alternative staff scheduling patterns. Patient flow is particularly sensitive to the shift scheduling of physicians.

A systems analysis and model of driver licensing in the state of Florida

The structure of the system that licenses drivers in the State of Florida is addressed. A SLAM simulation model of the system is employed to study various working patterns in the multiple-station multiple-queue service centers that provide licenses. The theory of such systems is addressed and alternate, efficient working patterns suggested. (Simulation, Queuing, Work Planning, Scheduling )

A systems analysis and model of real-time skin cancer treatment

Savings in health care costs, system efficiency, and improved patient satisfaction accruing from a real-time (simultaneous) approach to treatment of skin cancer is examined. The hypothesized benefits of a real-time system design are estimated using a SLAM based simulation model of a typical dermatology clinic specializing in the treatment of basal cell and squamous cell carcinomas.

A soft systems approach to input distribution estimation for a non-stationary demand process

A soft systems methodology is undertaken to develop a design for a communications system to support a political campaign. The focus is on the methodology to estimate an input distribution for a SLAM-based simulation model. Two problems-the lack of referent system data and the expectation of non-stationary system demand-indicate a need for a soft systems approach. The results of this study suggest that the approach is effective for estimating an input distribution for simulation analysis. The case provides an example of the contrast between the soft systems and hard systems approaches.

A systems analysis and model of a parallel multi-server queueing system

A feasible set of management policies to handle multitandem queuing systems characterized by stochastic demand is presented. This type of system is common in many service related organizations. The research used a previously developed simulation model of a State of Florida driver licensing office to manipulate policy variables in an effort to provide options for increased system efficiency. A key consideration is the ability to balance customer waiting time against ideal server time. The use of simulation analysis enables the modeler to incorporate complex characteristics of this type of system into the model, therefore providing a realistic representation of possible management actions. Dramatic results produced by the moving server models indicate that this method may provide the most efficient pairing of server resources with customer demand.<<ETX>>