Public Policy Group Support Systems Part 1

1.Research Motives and Objectives

In the wake of a changing, international community, public policy has become a major topic of discussion for governments when considering present and future undertakings. The ever increasing complexity of public affairs, and rapid expansion in how a government functions, brings a diverse number of problems policy makers must face, and, through no lack of effort on their part, its not possible for them to take everything into consideration. In order to supplement bias shown by the individual, I hope to integrate systems implementing expert knowledge and information technology to support policy makers in arriving at optimum decisions in an environment full of risk and uncertainty.

Future society will be an information society, and Japanese futurologist Zeng-Tian Mi-Er in his book "Managing In the Information Society", predicts the future information society will be an equal society of citizen participation, information sharing and division of labor (translated by You Wan-Juan, 1984). In pace with the integration and universal application of computers and communication technology, a nation’s overall strength will rely heavily on its ability to put information to good use. President Kennedy's, political advisor, T.C. Sorensen, once said: "Intellectual Interaction between many people far exceeds an individual’s intuition in inspiring thoughts (Sorensen, 1963:59)." 1978 economics Nobel prize winner Simon, in his momentous work ”Administrative Behavior”, considered modern government policy to be nothing but a product of "group consultation" (Simon, 1976). Therefore, the main objective of this dissertation is to conceive a ”Group Support System” workable in a public policy decision process. Hopefully, this will promote communication, coordination, information striking a common objective - supporting policy makers formulating policy alternatives, consolidating policy analysis skills, enhancing policy efficiency and quality, finding the optimum government policy and solving policy problems.

2. Research Framework and Methods

1. Research Framework

The framework of this research paper can be looked at from two perspectives: the vertical perspective defines the public policy decision process into several divisions — form of policy problem, policy planning policy evaluation and policy implementation; the horizontal perspective examines the problem by relating several different functional modules. The operating system is thus designed with this framework in mind, see Diagram 1.

2. Research Methods

The research methods used in mis dissertation are listed as follows:

(1)Referenced Literature

Major decision theory models related to public policy making and research models on group support systems. Analysis of research developments on related group support systems and motivation from later Researchers

(2) Computer Simulation

Using computer simulation, analyzing how group support systems can be put to use in the process of public policy decision making — the conditions supporting policy making at every stage in the group support system from the form of policy problem, policy planning, policy implementation to policy assessment. The "Public Policy Making Group Support System" formulated in this research paper was established by carrying out analysis on, via computer simulation, sorting, statistics compiling and data recording/filing of data collections, classification coding, data and knowledge bases, etc.

(3) Analytical Hierarchy Process

The major function of Analytical Hierarchy Process (AHP) is to solve complex policy problems. As proposed by US Operations Research specialist, Thomas L. Saaty, in 1971, AHP is a system decision analytical model, mainly employed where multi-criteria decisions apply (Saaty, 1980). The steps involved in its application are : (1) define the problem, (2) identify the criteria, (3) structure the hierarchy, (4) construct a pair-wise comparison matrix, (5) obtain all judgment, (6) calculate priority vectors value and overall priority vectors, (7) evaluate consistency.

(4) Multi-attribute Utility Analysis

Multi-attribute Utility Analysis (MAUT), a resources distribution and decision making method for policy alternatives, originating from the "Bayesian Decision Method". Theoretical policy alternatives have the following basic attributes:

(a) Comparison between policy alternatives
(b) Any policy must take responsibility for citizens or diversified groups and their different benefits.
(c) Policy alternatives generally have multi-objectives, with different levels of importance affixed to each objective.
(d) Inter-policy alternatives must make subjective judgments.
(e) Subjective judgment policy, policy alternatives and objective scaling are indicated by means of number or position.
(f) Assessment results of a subjective judgment becomes a choice for making a policy decision (Ke San-Ji, 1991).

3. System Analyses and Design

Similar to building a house, the construction of any information system is a complex and substantial undertaking, and, similarly, needs a variety of tools, technology and scheduled operations plan in order to complete. Implementing the development of an information system requires a decision on the most suitable and efficient methods to deploy, and normally includes the utilization of software engineering. Research advancement in this area has given information system developers the following courses of research to follow:

(a) System analysis attacks the cmx of a problem by researching its cause, moreover. it ascertains the system's objective, requirements, handling procedure and operating methods, whereby it drafts various solutions to solving the problem. The major tool adopted in this phase is the data flow diagram (DFD), providing a general outline of the information system, utilizing data structure to set forth the objective's attributes, and making use of algorithms to detail each operating process.

(b) System Design: The objective of systems analysis is to decompose a large problem into a group of related individual bodies, thereby understanding the central problem. While the objective of system design is to assemble the work process, that is, researching on how to design the computer system so that it produces several smaller programs, then combining these into a working module that is capable of completing the prescribed objective. Appropriate use of structured design techniques can prolong the program's usefulness. Moreover, making use of modular programming helps in applying both singular and comprehensive testing of the system, as well as establishing a more coherent program basis.

(c) System implementation - Assessment criteria that must be considered when the system is running are: how to meet schedule requirements, budget requirements and performance requirements (Rosenau, 1986). The most common tool used in mis phase is Project Management.

Apart from the three phases described above, from an information management system's point of view, some scholars hold the opinion that a fourth phase should be added: installation and training, in order to deliver realistic results, support an organization's activities and achieve objectives (Chang Feng-Hsiung, 1990)

1System Analysis

The author used the above described software engineering development process to conceive a realistic group support information system methodology, and, from a system analysis perspective, interpret the system's central principle. The system operates through two major elements: the vertical element has as its basis the three main phases - form of policy problem, policy planning, policy implementation and evaluation. Each phase operates, according to function or module requirements, on a parallel plane in assisting the database, knowledge and rules base, inference engine or control procedure

From a modularization perspective, system operation flow composes of the following modules:

(1)Public Policy Issue Base Module

The issue base module is one addressed in the earlier handling stages (the other one being Module 2 - Expert Personal Database), responsible for pre-coding, classification and the storing of public policy issues in the issue database. The coding and database are focal points of the system design. Coding handles classification, problem definition and serial numbering, further processing is then carried out in me database — registering, categorizing, compiling statistics and storage. The optimum time for utilizing the issue base is when collecting information from the issue database, re-implementing an inference or deciding on a new policy, see diagram 2.

(2) Expert Personal Data Base Module

This is the second module addressed in the earlier handling stages of the whole operating system, its two major operations being: collecting and categorizing experts' and scholars' personal data, and the translation and registering of all opinions expressed by experts and scholars. The former operation is important in designing the database system, the latter in translation and establishment of a knowledge base system, see diagram 3.

(3) Group Inference Module

This module holds the key to the whole system. A list of experts or scholars, most suited to participate in a specifically designated issue from module 1, is established by the module 2 database. A filtering process is carried out, and the expert name list indexed. From the knowledge base (established by module 2), all related viewpoints expressed by these experts are sought and displayed. Using these opinions as reference, through further singling out by researchers or knowledge engineers, a list of the experts' opinions is made. The computer collates the expert's opinions and approves the right persons for the issue involved, see diagram 4.

(4) Group Decision Module

After collecting and sorting the opinions obtained from module 3, the analytic hierarchy process takes affect, deciding on each opinion's level of relative importance, this becoming source criteria for policy makers, staff officers or reference material. The focal point c made from the knowledge base the researchers to reference in the design of policy alternatives. Finally, in order to meet the policy alternative's objective, multi-attribute utility analysis is employed in selecting the optimum policy alternative, providing policy makers with important reference material. The focal point of this module's design is its outline of selections made from the knowledge base, the AHP computing process and its compilation of statistical results, see diagram 5

(5) Monitoring and Learning Module

Once a policy is selected, this module's first responsibility is handling the input of implementation assessment criteria for this policy, this acting as reference for monitoring purposes. This phase merely records the file on disc. The module's next responsibility, during policy implementation, is the periodic or non-periodic input of policy implementation feedback data and the saving of this data on disc. Finally, a comparison is made between the predicted policy objective and actual accomplished objective, compiling any discrepancies found. This is then translated and stored in the knowledge base, becoming reference material for making decisions on future, related issues.