Crime Management System Dbms - DOC by qov41983


Crime Management System Dbms document sample

More Info
									                                                    Unit 3
                                          Decision Support Systems

Discussion in DSS relate to

The two foci, efficiency and effectiveness
Monitoring in DSS modelling processes
Decision Rules
Decision tables as an example of the creation of decision rules
The Nature of DSS
Characteristics of DSS
DSS Benefits
Typology of DSS
DSS and Personnel Roles

Relates to:
 the quantity of outputs which can be generated
 the means by which a given defined task can be performed, in order to achieve outputs as well as possible.

 is quantitative and can be calculated by the amount of output which can be generated for every unit of
 implies a narrowing of focus to get a job done. e,g., minimising time, cost, or effort in order to complete an
     activity given a constant output, or maximising output for a constant input.

Efficiency is usually connected with MIS, the production of outputs from the system applying criteria of

 to the quality of accomplishment
 means of accomplishment

 what should be done qualitatively
 criteria influencing a decision which will impact on an organisational goal.
 the effect of decisions or approaches on events which directly impinge on the degree of accomplishment in
   achieving the organisational objectives, implying a broadening of focus in order to find what set of
   activities to take into account.

Effectiveness is typically connected to DSS, for it is within such a servicing environment that mechanisms can
be applied which improve the quality of decisions.

Another aspect of DSS is structure.

This is the relationship between definable entities like people, objects, or processes.
A structural relationship can be:
 static, when it will not change over time
 dynamic, when it changes over time

   well structured, when the entities and their mutual relationships are known, and causal relationships can
    be defined in a sort of cause-effect chain
   unstructured or illstructured that involve entities that do not have a relationship, either across the enti-
    ties or between them across time

Structured is dependent on knowledge and understanding.
Two forms of structure:
 deep
 surface
(Keen and Scott Morton, 1978, p93; Chomsky, 1975)

Surface structure:
 entities operate to carry out their functions or purpose
 relates to behaviour of the set of entities as a whole

Deep structure:
 entities have group coherence that gives (deep) meaning to the entities as a group
 dependent on worldview of an inquirer

Types of Structure in Decision Processes

Thrre types of structuring in decision making:
 unstructured,
 structured,
 semi-structured.

      Characteristic of decision making                        Unstructured                          Structured
Objectives                                        Ambiguous and non-operational, or      Unambiguous and operational, and
                                                  relatively operational but numerous    are non-conflicting
                                                  and conflicting.
Retrospective cause of changes in decision        Difficult and non-deterministic        Straightforward and deterministic
outcomes and prediction of decision outcomes of
the actions taken by the decision maker
What actions taken by the decision maker might    Uncertain                              Certain
effect decision outcomes
Advanced knowledge of data requirements           No. Data retrieval must allow for ad   Yes
                                                  hoc retrieval requests
Programmable decision processes                   No                                     Yes

A decision process which has both elements of structured and unstructured decision process are said to be

                         Programmable and Non-programmable Decision Processes

                                          Structured DP

                                                   Semi-structured DP

                                                       Unstructured DP

             Relationship between Unstructured, structured and Semistructured Decision Processes

     Type of Decision Process                                            Explanation

Programmable                         Able to be prespecified by a set of rules or DP decision procedures; implies
                                     certainty since set of outcomes must be known.
Nonprogrammable                      No pre-established decision rules or procedures, DP     in situations of flux

                Relationship between programmable and non-programmable decision processes

                                     high             Non-programmable DP


                                                      Programmable DP

Decision Rules
 determined by identifying the criteria necessary for making a decision.
 involves routine programmable components
 when parts of a subsystem of an organisation are well known, then decision rules can be established, and
    different parts of the organisation can operate without feedback
 when monitoring suggests that an atypical situation has developed, then feedback to a decision maker is
    required so that a decision rule can be adjusted or a new one introduced temporarily.

E.g., domestic resources such as electricity or gas will have decision rules concerned with the issue of accounts
each quarter.

Decision Tables as an example of Decision Rules

Initial Actions (IA): consist of actions which are unconditionally performed before any conditions have been

Conditions (C): y represents a yes, n represents a no, a full stop . represents a null consideration, and an * is a
wild card symbol that means this condition is immaterial for this rule.

Action (A): x means perform an entry, and a full stop . means do not perform it

 IA      Get up                                   ─┐
         Get dressed                               │ Initial Actions
         Have breakfast                           ─┘
                             1 2 3 4 5 6
 C       Is it raining       y y n n n n                      ─┐
         Weather forcast fine* * y y n n                       │conditions
         Is it warm today    n y y n y n                      ─┘
 A       Take raincoat       . x . . . .                      ─┐
         Take umbrella       x . . . x x                       │actions
         Take overcoat            x . . x                      . x    │
         Take none of above . . x . . .                       ─┘

Which means:
R1:     IF      it is raining AND it is not warm today
        THEN take an umbrella AND take an overcoat.

R2:     IF          it is raining AND it is warm today
        THEN        take a raincoat
R3:     IF          it is not raining AND the weather forcast is fine AND it is warm today
        THEN        do not take an umbrella, a raincoat, or an overcoat
R4:     IF          it is not raining AND the weather forcast is fine AND it is not warm today
        THEN        take an overcoat
R5:     IF          it is not raining AND the weather forcast is not fine AND it is warm today
        THEN        take an umbrella
R6:     IF          if is not raining AND the weather forcast is not fine AND it is not warm today
        THEN        take an umbrella AND take an overcoat

Monitoring in DSS modelling processes

                                          Decision Process with Monitoring

                                       Data Input

    Decision Rule                     Program or                       Monitoring
      Options                         Algorithm

                                                          programmed            no
                                                         circumstances                     Automatic
              Feedback                                                        selection of
             to Manager                                                                  possible
                                                                yes                      options

                                                                                               Select &

Two types of monitoring common:
 variance monitoring about some issue such as financial soundness
 binary monitoring (its occurrence or non-occurrence) about an issue such as debt

Other Attributes of Decision Processes

The DSS seeks to increase the decision maker's ability to deal with complexity and uncertainty, and can aid

Examples of use of DSSs:
1.     Strategic Planning: Decisions are related to setting policies, choosing objectives, and selecting
2.     Management Control: Decisions are related to assuming effectiveness in acquiring and using
3.     Operational Control: Decisions are related to assuming effectiveness in performing operations
4.     Operational Performance: Decisions that are made while performing the operations

Examples of the relationship between types of decisions and structure are given in the table below:

   Decision Process           Operational                    Operational                    Management         Strategic Planning
                             Performance                       Control                        Control
      Structured          Payroll Production              Accounts receivable            Budget analysis       Plant location
    (MIS support)
 Semi-structured (DSS     Dispatching                     Production                     Forecasting           Mergers
       support)                                           scheduling
 Unstructured (human      Solving a crime                 Financial                      Budget reports        Product planning
     interaction)                                         management

                                                     The Nature of DSS

                                                            Input               Decision support

                                                   data                          Database          Modelling
                                      feedback                                                      base

                        Environment            System of
                                               operations                                   Analysis

                                                                       of situation
                                          Output            decision

                                 Context diagram for a decision support system

Characteristics of DSS
DSS can be used as an early warning system. They for instance can highlight the following:

 interactive computer-based systems
 support decision making
 decisions involving judgement
 database and network requirement
 knowledge base especially for novice decision makers to learn about decision processes
 simulation to develop experiential knowledge about complex situations

Components of a DSS
 Control processes especially including monitoring
 inquiry through:
       analysis of situation and its context,
       synthesis of alternative decision options, and
       choice of decision options that will formulate a decision.

Early warning system
 pattern exception
 adaptive learning system
 transient and sustained change
 operational and commercial significance
 early awareness of threats and opportunities
 automatic surveillance
 management by exception

DSS Benefits
 Concentrates time on important issues
 early warning time to decide on action
 proactive management instead of reactive
 system encourage best business practices
 data availability on all factors in the decision
 user selected automatic analysis of data
 knowledge base recommendations
 encourages user to make inquiry
 ability to adjust basis of decisions to other inquirers

Typology of DSS
Three types of DSS:
 Specific DSS
 DSS generators
 DSS tools

Specific DSS
Two types:
 an MIS application and could be classed as a stand-alone MIS add-on, with characteristics which
    distinguish it from a typical data processing application (e.g., a portfolio management system, and a police
    beat allocation system
 a stand alone systems not connected to an organisational MIS (e.g., knowledge based systems

DSS Generator
This is a package that enables specific DSS aids to be created. An example of such a package is the GADS
(Geodata Analysis and Display System) that is essentially an advanced 4GL in that it satisfies end-user
requirements by being able to relatively quickly integrate:
                             Menus, maps, data, choices, procedures, commands.

Another DSS generator is the EIS (Executive Information System) which integrates the activities of:
 report generation, report preparation, inquiring capability, modelling language, graphic display commands,
                                 financial and statistical analysis subroutines.

DSS Tools
Enable the generation of. Fourth generation operating systems and special purpose languages that can create
DSS generators and specific DSS.

DSS and Personnel Roles
The three levels of DSS application can be associated with five personnel roles, as shown below:

              Levels of DSS Application                                      Personnel Roles

                     Specific DSS                                           Decision Maker
                                                                          Decision Making Staff
                    DSS Generator                                             DSS Builder
                                                                           Technical Support
                      DSS Tools                                                 Toolsmith

Sprague's idea of the three levels of development in DSS require one of the two options:
1.       the DSSs are contained within a large organisation that can afford all levels of development
2.       specific DSSs are made commercially available as a Lego type matrix of small systems
         which can be plugged together to generate more complex and comprehensive systems,
         typically sold to smaller firms without large resources; in this model DSS tools could be the
         domain of consultancy firms which could generate specific or integrated DSS as required.

A DSS Vision
The DSS developers vision of a general system which can enable bolt-on compatible packages for the end user
to easily develop sophisticated DSS, in the same way as modern integrated packages connect spreadsheet,
database, wordprocessors, and statistical package, is still some way off. At the lower end of the cost spectrum,
these integrated packages do at least currently provide a facility of what may be classed as a rudimentary DSS.

                                     DSS, Modelling and Data Capture

                                      Base elements
                                       of a DSS

                                database       modelbase             knowledgable
                               subsystem       subsystem                user

                                      DBMS    MBMS


                                    Relationship between DSS Subsystems

   DGMS - is a dialogue generation and management system which enables the user to use the
    system in a user oriented way
   MBMS - a model base management system which enable models to be collected, assessed for
    suitability according to their defined applications domains, and applied
   DBMS - a database management system as normally contained within an MIS.

                                           DSS Database Subsystem
DSS should:

enable access to distributed database systems through an appropriate DBMS and have characteristics which are
in common with a distributed DBMS.
According to Sprague Jr. the following characteristics are necessary for a DSS database subsystem:
 the ability to combine a variety of data sources through a data capture and extraction process
 the ability to add and delete data sources quickly and easily
 the ability to portray logical data structures in user terms so that the user understands what is
     available and can specify needed additions or deletions
 the ability to handle personal and unofficial data so that the user can experiment with alternatives
     based on personal judgement
 the ability to manage the wide variety of data with full range of data management functions.

The functions of the Decision Support Database are:
         1)        create and generate data
         2)        restructure data
         3)        update
         4)        inquiry and retrieval

                                Representation of the Database Subsystem

          Database access to:
            External data source
            Functional data of the organisation and application data
            Other internal data sources

          Data processor:
             Extraction                                         Decision Support Database

            Model Base                             Management System and user interface

                                              The Model Subsystem
Sprague Jr. proposes the following characteristics of a model subsystem:
 the ability to create new models quickly and easily
 the ability to catalog and maintain a wide range of models, supporting all levels of management
 the ability to interrelate these models with appropriate linkages through the database
 the ability to manage model base management functions in a way which is equivalent to DBMS:
   that is involving storage, cataloging, linking, retrieving.

                                 Representation of Modelling Subsystem

                                                       Model Base:
                                                        Strategic, tactical
             Database                                   operational models

                                                          Model Building blocks
                                                            and subroutines

             DSS Management
             User interface

                               Data Handling and Model Building in DSSs

   requires that data capture and retrieval from a DBMS is sufficiently flexible to enable rapid modifications
    to occur in response to ad hoc user requirements.

   should be able to integrate data retrieval with decision models, whether the latter are normative optimising
    models, data flow, graphical or generally descriptive.
   models which are used should be appropriate for the class of data available, and it should not necessarily
    be part of a decision maker's staff knowledge to determine the compatibility between problem situations
    and models adopted.
   More recently relational database theory has been applied to model bases, involving such rules as join,
    select, project.


To top