MBA 544-Advanced Decision Support System

8/9/2019 MBA 544-Advanced Decision Support System

1/171

Punjab Technical University

World over distance Education is fast growing mode of education because of the unique benefitsit provides to the learners. Universities are now able to reach the community which has for solong been deprived or higher education due to various reasons including social, economic andgeographical considerations. Distance Education provides them a second chance to upgrade theirtechnical skills and qualifications.

Some of the important considerations in initiating distance education in a country like India, has been the concern of the government in increasing access and reach of higher education to a largerstudent community. As such, only 6-8% of students in India take up higher education and morethan 92% drop out before reaching 10+2 level. Further, avenues for upgrading qualifications,while at work, is limited and also modular programs for gaining latest skills through continuingeducation programs is extremely poor. In such a system, distance education programs providethe much needed avenue for:

Increasing access and reach of higher education;

Equity and affordability of higher education to weaker and disadvantaged sections of thesociety;

Increased opportunity for upgrading, retraining and personal enrichment of latestknowledge and know-how;

Capacity building for national interests.

One of use important aspects of any distance education program is the learning resources.Learning material provided to the learner must be innovative, thought provoking,comprehensive and must be tailor-made for self-learning. It has been a continuous process for theUniversity in improving the quality of the learning material through well designed coursematerials in the SIM format (self-instructional material). While designing the material, theuniversity has researched the methods and process of some of the best institutions in the worldimparting distance education.

About the University

Punjab Technical University (PTU) was set up by the Government of Punjab in 1997 through astate Legislative ACT. PTU started with a modest beginning in 1997, when University had onlynine Engineering and thirteen Management colleges affiliated to it. PTU now has affiliated43 Engineering colleges, 56 colleges imparting Management and Computer Application courses,20 institutions imparting pharmacy education, 6 Architecture institutions, 2 Hotel Management


2/171

and 12 Regional Centres for imparting M. Tech and Ph. D Programs in different branches ofEngineering and Management. During a short span of nine years, the University has undertakenmany innovative programs. The major development during this period is that University hasrestructured its degree program and upgraded syllabi of the course in such a way as to increasethe employability of the student and also to make them self-reliant by imparting HigherTechnical Education. We at Punjab Technical University are propelled by the vision and wisdomof our leaders and are striving hard to discharge our duties for the overall improvement ofquality of education that we provide.

During a short span of nine years, the University has faced various challenges but has alwayskept the interest of students as the paramount concern. During the past couple of years, theUniversity has undertaken many new initiatives to revitalize the educational programs impartedwith the colleges and Regional centers.

Though knowledge and skills are the key factors in increasing the employability and competitiveedge of students in the emerging global environment, an environment of economic growth andopportunity is necessary to promote the demand for such trained and professional manpower.The University is participating in the process of technological growth and development inshaping the human resource for economic development of the nation.Keeping the above facts in mind Punjab Technical University, initiated the distance educationprogram and started offering various job oriented technical courses in disciplines like informationtechnology, management, Hotel Management, paramedical, Media Technologies and FashionTechnology since July 2001. The program was initiated with the aim of fulfilling the mandate ofthe ACT for providing continuing education to the disadvantaged economically backwardsections of society as well as working professionals for skill up-gradation.

The University has over the years initiated various quality improvement initiatives in running itsdistance education program to deliver quality education with a flexible approach of educationdelivery. This program also takes care of the overall personality development of the students.Presently, PTU has more than 60 courses under distance education stream in more than 700learning centers across the country.


3/171

About Distance Education Program of PTU

Over the past few years, the distance education program of PTU has gained wide publicity andacceptance due to certain quality features which were introduced to increase the effectiveness oflearning methodologies. The last comprehensive syllabus review was carried out in the year 2004-05 and the new revised syllabus was implemented from September 2005. The syllabus oncereviewed is frozen for a period of 3 years and changes, if any, shall be taken up in the year 2008.Various innovative initiatives have been taken, which has increased the popularity of theprogram. Some of these initiatives are enumerated below:1. Making a pyramid system for almost all courses, in which a student gets flexibility of

continuing higher education in his own pace and per his convenience. Suitable credits areimparted for courses taken during re-entry into the pyramid as a lateral entry student.

2. Relaxed entry qualifications ensure that students get enough freedom to choose theircourse and the basics necessary for completing the course is taught at the first semesterlevel.

3. A comprehensive course on Communications and Soft Skills is compulsory for allstudents, which ensures that students learn some basic skills for increasing theiremployability and competing in the globalized environment.

4. Learning materials and books have been remodeled in the self-Instructional Materialformat, which ensures easy dissemination of skills and self-learning. These SIMs are givenin addition to the class notes, work modules and weekly quizzes.

5. Students are allowed to take a minimum of 240 hours of instruction during the semester,which includes small group interaction with faculty and teaching practical skills in apersonalized manner.

6. Minimum standards have been laid out for the learning centers, and a full time counselorand core faculty is available to help the student anytime.

7. There is a wide network of Regional Learning and Facilitation Centers (RLFC) catering toeach zone, which is available for student queries, placement support, examination relatedqueries and day-to-day logistic support. Students need not visit the University for any oftheir problems and they can approach the RLFC for taking care of their needs.

8. Various facilities like Free Waiver for physically challenged students, Scholarship scheme by the government for SC/ST candidates, free bus passes for PRTC buses are available tostudents of the University.

The university continuously aims for higher objectives to achieve and the success always gears usfor achieving the improbable. The PTU distance education fraternity has grown more than 200%during the past two years and the students have now started moving all across the country andabroad after completing their skill training with us.

We wish you a marvelous learning experience in the next few years of association with us!

DR. R. P. SINGHDean

Distance Education


4/171

Dr. S. K. Salwan

Vice Chancellor

Dr. S. K. Salwan is an eminent scientist, visionary and an experienced administrator. He is adoctorate in mechanical engineering from the IIT, Mumbai. Dr. Salwan brings with him 14 yearsof teaching and research experience. He is credited with establishing the Department of DesignEngineering at the institute of Armament Technology, Pune. He was the founder-member of theintegrated guided missile programme of defence research under His Excellency Honorable Dr.A.P.J. Abdul Kalam. He also established the high technology missile center, RCI at Hyderabad.He has been instrumental in implementing the Rs 1000-crore National Range for Testing Missilesand Weapon Systems at Chandipore, Balance in a record time of three years. He was director ofthe Armament Research and Development Establishment, Pune. Dr. Salwan has been part ofmany high level defence delegations to various countries. He was Advisor (Strategic project) andEmeritus Scientist at the DRDO. Dr. Salwan has won various awards, including the Scientist of

the Year 1994; the Rajiv Ratan Award, 1995, and a Vashisht Sewa Medal 1996, the TechnologyAssimilation and Transfer Trophy, 1997 and the Punj Pani Award in Punjab for 2006.

Dr. R.P. Singh

Dean, Distance Education

Dr. R.P. Singh is a doctorate in physics from Canada and has been a gold medallist of BanarasHindu University in M.Sc. Dr. Singh took over the Department of Distance Education inNovember 2004 and since then the University has embarked on various innovations in DistanceEducation.

Due to combined efforts of the department, the RLFCs and Centers, and with active support ofthe Distance Education Council headed by Dr. O.P. Bajpai, Director University College ofEngineering Kurukshetra University the distance education program of PTU is now a structuredsystem which empowers the learner with requisite skills and knowledge which can enhance theiremployability in the global market. Dr. R. P. Singh is promoting distance education at thenational level also and is a founder member of Education Promotion Society of India and ismember of various committees which explores innovative ways of learning for the disadvantagessections of society. The basic aim of the distance education programs has been to assimilate allsections of society including women by increasing the access. Reach, equity and affordability ofhigher education in the country.


5/171

ADVANCED DECISION SUPPORT SYSTEMS

MBA-544

This SIM has been prepared exclusively under the guidance of Punjab Technical University (PTU)and reviewed by experts and approved by the concerned statutory Board of Studies (BOS). Itconforms to the syllabi and contents as approved by the BOS of PTU.


6/171

Copyright Anindita Hazra, 2008

No part of this publication which is material protected by this copyright notice may bereproduced or transmitted or utilized or stored in any form or by any means now known orhereinafter invented, electronic, digital or mechanical, including photocopying, scanning,recording or by any information storage or retrieval system, without prior written permissionfrom the publisher.

Information contained in this book has been published by Excel Books Pvt. Ltd. and has beenobtained by its authors from sources believed to be reliable and are correct to the best of theirknowledge. However, the publisher and its author shall in no event be liable for any errors,omissions or damages arising out of use of this information and specifically disclaim any impliedwarranties or merchantability or fitness for any particular use.

Published by Anurag Jain for Excel Books Pvt. Ltd., A-45, Naraina, Phase-I, New Delhi-110 028Tel: 25795793, 25795794 email: [email protected]


7/171

PTU DEP SYLLABI-BOOK MAPPING TABLE

MBA-544 ADVANCED DECISION SUPPORT SYSTEMS

Syllabi Mapping in Book

Section I

Decision-making: Concept, Process, Simons model,Programmed versus non-programmed decisions, quantitativetools.

Decision-models: Decision making under assumed certainty, risk& uncertainty.

Introduction to DSS: Characteristics and Objectives: Comparisonwith EDP/MIS.

Levels of Decision Support System: Specific, Generator and tools Forecasting packages, Statistical packages; Relationship

Section II

Role of Decision Support Systems and its application.

Components of Decision support Systems: Data Subsystem,

Model Subsystem, and User-interface,. DBMS, Quantitativemodels and modeling in DSS.

Group Decision support Systems, Expert system and itsintegration with DSS. Executive Support System.

Section III

Create Applications Using EXCEL

Data Warehousing: Concepts, database structure,

Getting Data into the Data Warehouse.

Data Mining: Automated Analysis, constructing a datawarehouse system.

Unit 1: Decision-Making(Page 3-16)

Unit 2: Decision-Models(Page 17-21)

Unit 6: Components of DSS

(Page 71-78)

Unit 5: Role of DSS and itsApplication(Page 63-69)

Unit 8: Create Applicationsusing EXCEL (Page 87-104)

Unit 3: Introduction to DSS(Page 23-31)

Unit 4: Levels of DecisionSupport System

(Page 33-60)

Unit 7: Subclasses of DSS(Page 79-84)

Unit 9: Data Warehousing

(Page 105-120)Unit 10: Getting Data into the

Data Warehouse (Page 121-141)

Unit 11: Data Mining (Page 143-165)


8/171

ContentsSection-I

UNIT 1 DECISION-MAKING 3

IntroductionConcept of Decision-MakingDecision-Making ProcessDecision-Making is a Recursive ProcessSimons ModelProgrammed versus Non-Programmed DecisionsQuantitative ToolsSummaryKeywordsReview QuestionsFurther Readings

UNIT 2 DECISION-MODELS 17IntroductionDecision-Making TheoryDecision-Making under Assumed CertaintyRisk and Uncertainty of Decision MakingSummaryKeywordsReview QuestionsFurther Readings

UNIT 3 INTRODUCTION TO DSS 23IntroductionConcept of DSSCharacteristics and Objectives of DSSAdvantage and Limitation of Decision Support SystemComparison with EDP/MISSummaryKeywordsReview QuestionsFurther Readings

UNIT 4 LEVELS OF DECISION SUPPORT SYSTEM 33

IntroductionLevels of Decision Support SystemForecasting PackagesStatistical PackagesRelationshipSummary


9/171

KeywordsReview QuestionsFurther Readings

Section-II

UNIT 5 ROLE OF DSS AND ITS APPLICATION 63IntroductionRole of DSS and its ApplicationSummaryKeywordsReview QuestionsFurther Readings

UNIT 6 COMPONENTS OF DSS 71IntroductionComponents of Decision Support SystemsData Sub-systemModel Sub-systemUser-interfaceDBMSQuantitative ModelsModeling in DSSComponents of Decision ModelsSummaryKeywordsReview QuestionsFurther Readings

UNIT 7 SUBCLASSES OF DSS 79IntroductionGroup Decision Support SystemsExpert System and its Integration with DSSExecutive Support SystemSummaryKeywordsReview QuestionsFurther Readings

Section-III UNIT 8 CREATE APPLICATIONS USING EXCEL 87

IntroductionConcept of Excel ProgrammingCharacteristic Features of a Spreadsheet ApplicationAdvantages of Excel ProgrammingProgramming FundamentalsThe VBA Language


10/171

Excel Object ModelMacros and ProgrammingDesigning Your Custom ApplicationYour First Excel ProgramDeveloping Applications, Step by StepSummaryKeywordsReview QuestionsFurther Readings

UNIT 9 DATA WAREHOUSING 105IntroductionConcepts of Data WarehousingAdvantages of using Data WarehouseDatabase StructureData Warehouse ArchitecturesDeveloping a Data Warehouse Architecture

Developing Data WarehousesDeveloping a Data Warehouse StrategyDesigning Data WarehousesManaging Data WarehousesFuture DevelopmentsSummaryKeywordsReview QuestionsFurther Readings

UNIT 10 GETTING DATA INTO THE DATA WAREHOUSE 121IntroductionGetting Data into the Data WarehouseData PreprocessingData PreparationData CleaningData Integration and TransformationChallenges of Data IntegrationOrganizational ChallengesData ReductionDiscretization and Concept HierarchySummaryKeywordsReview QuestionsFurther Readings

UNIT 11 DATA MINING 143IntroductionDefinition of Data MiningRole of Data MiningImportance of Data MiningData Mining in Business


11/171

Historical Evolution of Data MiningWorking of Data MiningAutomated AnalysisConstructing a Data Warehouse SystemData Warehouse FrameworkSummaryKeywordsReview QuestionsFurther Readings


12/171

Unit

Decision-Making

Unit 2

Decision-ModelsUnit 3

Introduction to DSS

Unit 4

Levels of Decision Support System

SECTION-I


13/171

Decision-Making

Notes

Punjab Technical University 3

Unit 1 Decision-Making

Unit Structure Introduction Concept of Decision-Making Decision-Making Process Decision-Making is a Recursive Process Simons Model Programmed versus Non-Programmed Decisions Quantitative Tools Summary Keywords Review Questions Further Readings

Learning Objectives After studying this unit you will be able to understand:

Concept of Decision-Making Overview of Decision-Making Process Brief idea about Simons Model Differences between Programmed versus Non-Programmed Decisions Quantitative Tools like decision tree, payback analysis and simulation

IntroductionWe all make decisions of varying importance every day, so the idea that decisionmaking can be a rather sophisticated art may at first seem strange. However, studieshave shown that most people are much poorer at decision making than they think. Anunderstanding of what decision making involves, together with a few effectivetechniques, will help produce better decisions.

Decision-making is about facing a question, such as, "To be or not to be?", i.e., to bethe one you want to be or not to be? That is a decision. A decision usually involvesthree steps:

A recognition of a need: A dissatisfaction within oneself - a void or need; A decision to change - to fill the void or need; A conscious dedication to implement the decision.

So aside from that, we see that making the correct decisions is not only what we wantto do, but includes what we have to do.

In this unit we shall discuss about decision-making.

Concept of Decision-MakingDecision making can be defined as the study of identifying and choosing alternatives

based on the values and preferences of the decision maker. Making a decision impliesthat there are alternative choices to be considered, and in such a case we want not


14/171

Advanced DecisionSupport Systems

Notes

4 Self-Instructional Material

only to identify as many of these alternatives as possible but to choose the one that best fits with our goals, desires, lifestyle, values, and so on.

It may also be defined as the process of sufficiently reducing uncertainty and doubtabout alternatives to allow a reasonable choice to be made from among them. Thisdefinition stresses the information gathering function of decision making. It should benoted here that uncertainty is reduced rather than eliminated. Very few decisions aremade with absolute certainty because complete knowledge about all the alternativesis seldom possible. Thus, every decision involves a certain amount of risk.

There are several basic kinds of decisions.:

1. Decisions whether: This is the yes/no, either/or decision that must be made before we proceed with the selection of an alternative. Should I buy a new TV?Should I travel this summer? Decisions whether are made by weighing reasonspro and con. The PMI technique discussed in the next unit is ideal for this kind ofdecision.

It is important to be aware of having made a decision whether, since too often weassume that decision making begins with the identification of alternatives,assuming that the decision to choose one has already been made.

2. Decisions which: These decisions involve a choice of one or more alternativesfrom among a set of possibilities, the choice being based on how well eachalternative measures up to a set of predefined criteria.

3. Contingent decisions: These are decisions that have been made but put on holduntil some condition is met.

Fig 1: Complex Problem Solving Situation

For example, I have decided to buy that car if I can get it for the right price; I havedecided to write that article if I can work the necessary time for it into my schedule.

Most people carry around a set of already made, contingent decisions, just waiting forthe right conditions or opportunity to arise. Time, energy, price, availability,opportunity, encouragementall these factors can figure into the necessary conditionsthat need to be met before we can act on our decision.


15/171

Decision-Making

Notes


Decision-Making ProcessProblem solving and decision-making are important skills for business and life.Problem-solving often involves decision-making, and decision-making is especiallyimportant for management and leadership. There are processes and techniques toimprove decision-making and the quality of decisions.

Decision-making is more natural to certain personalities, so these people should focusmore on improving the quality of their decisions. People that are less natural decision-makers are often able to make quality assessments, but then need to be more decisivein acting upon the assessments made.

Problem-solving and decision-making are closely linked, and each requires creativityin identifying and developing options, for which the brainstorming technique isparticularly useful. SWOT analysis helps assess the strength of a company, a businessproposition or idea; PEST analysis helps to assess the potential and suitability of amarket.

Good decision-making requires a mixture of skills: creative development andidentification of options, clarity of judgement, firmness of decision, and effective

implementation. For group problem-solving and decision-making, or when aconsensus is required, workshops help, within which you can incorporate these toolsand process as appropriate.

The six steps of this natural, intuitive decision-making process are: Step 1: Define the problem Step 2: Identify available alternative solutions to the problem Step 3: Evaluate the identified alternatives Step 4: Make the decision Step 5: Implement the decision Step 6: Evaluate the decision

Fig 2: Six Step Decision Making Process


16/171


Notes


Step 1: Define the problem

The most significant step in any decision making process is describing why a decisionis called or and identifying the most desired outcome(s) of the decision makingprocess.

One way of deciding if a problem exists is to couch the problem is terms of what onewanted or expected and the actual situation. In this way a problem is defined as thedifference between expected and/or desired outcomes and actual outcomes.

This careful attention to definition in terms of outcomes allows one to clearly state theproblem. This is a critical consideration because how one defines a problemdetermines how one defines causes and where one searches for solutions.

The limiting aspect of the problem definition step is not widely appreciated. Considerthis example.

Your company owns an old, downtown office building. Tenants are complaining thattheir employees are getting angry and frustrated because there is always a long delaygetting an elevator to the lobby at rush hour.

You are asked for a reaction on how to solve this problem. As with most problemsituations there are several ways to define the situation and several solutions thatsuggest themselves.

In this scenario the most common alternatives were: Flexible hours- so all the tenants' employees wouldn't be at the elevators at the

same time. Faster elevators - so each elevator could carry more people in a given time

period. Bigger elevators - so each elevator could carry more people per trip. Elevator banks- so each elevator would only stop on certain floors, increasing

efficiency.

Better elevator controls - so each elevator would be used more efficiently. More elevators - so that overall carrying capacity could be increased. Improved elevator maintenance - so each elevator would be more efficient. Encourage employees to use the stairs - so fewer people would use the elevators. If you examine each alternative you will see that several different definitions of

the problem must have existed.

If the solution is "flexible hours" the problem must have been defined as, "Too manypeople getting off work at a given time." No other problem makes sense for thatsolution.

"Faster elevators" comes from, "The elevators are too slow." "Bigger elevators" comes from, "The elevators are not carrying enough people." "More elevators" comes from, "Too few elevators."

The real life decision makers defined the problem as "people coming about having towait". Their solution was to make the wait less frustrating by piping music into theelevator lobbies. The complaints stopped.

There is no way that the eventual solution could have been reached if, for example,the problem had been defined as "too few elevators".

As you can see, how you define the problem determines where you go to look foralternatives/solutions, so define the problem carefully.


17/171

Decision-Making

Notes


Step 2: Identify available alternative solutions to theproblem

The key to this step is to not limit yourself to obvious alternatives or what has workedin the past but to be open to new and better alternatives. How many alternativesshould you identify? Ideally all of them. Realistically, we teach that the decisionmaker should consider more than five in most cases, more than three at the barestminimum. This gets away from the trap of seeing "both sides of the situation" andlimiting one's alternatives to two opposing choices; either this or that.

Step 3: Evaluate the identified alternatives

As you evaluate each alternative, you should be looking at the likely positive andnegative cones for each. It is unusual to find one alternative that would completelyresolve the problem and is heads and shoulders better than all others. Differences inthe "value" of respective alternatives are typically small, relative and a function of thedecision maker's personal perceptions, biases and predispositions.

As you consider positive and negative cones you must be careful to differentiate

between what you know for a fact and what you believe might be the case.The decision maker will only have all the facts in trivial cases. People alwayssupplement what facts they have with assumptions and beliefs.

This distinction between fact-based evaluation and non-fact -based evaluation isincluded to assist the decision maker in developing a "confidence score" for eachalternative. The decision maker needs to determine not just what results eachalternative could yield, but how probable it is that those results will be realized. Themore the evaluation is fact-based, the more confident he/she can be that the expectedoutcome will occur.

Step 4: Make the decision

When acting alone this is the natural next step after selecting the best alternative.When the decision maker is working in a team environment, this is where a proposalis made to the team, complete with a clear definition of the problem, a clear list of thealternatives that were considered and a clear rationale for the proposed solution.

Step 5: Implement the decision

While this might seem obvious, it is necessary to make the point that deciding on the best alternative is not the same as doing something. The action itself is the first real,tangible step in changing the situation. It is not enough to think about it or talk aboutit or even decide to do it. A decision only counts when it is implemented. As LouGerstner (CEO of IBM) said, "There are no more prizes for predicting rain. There areonly prizes for building arks."

Step 6: Evaluate the decision

Every decision is intended to fix a problem. The final test of any decision is whetheror not the problem was fixed. Did it go away? Did it change appreciably? Is it betternow, or worse, or the same? What new problems did the solution create? Thereforedecision-making is a recursive process.


18/171


Notes


Decision-Making is a Recursive ProcessA critical factor that decision theorists sometimes neglect to emphasize is that in spiteof the way the process is presented on paper, decision-making is a nonlinear,recursive process. That is, most decisions are made by moving back and forth

between the choice of criteria (the characteristics we want our choice to meet) and theidentification of alternatives (the possibilities we can choose from among).The alternatives available influence the criteria we apply to them, and similarly thecriteria we establish influence the alternatives we will consider. Let's look at anexample to clarify this.

Suppose someone wants to decide, Should I get married? Notice that this is a decisionwhether. A linear approach to decision making would be to decide this question byweighing the reasons pro and con (what are the benefits and drawbacks of gettingmarried) and then to move to the next part of the process, the identification of criteria(supportive, easy going, competent, affectionate, etc.).

Next, we would identify alternatives likely to have these criteria (Kantha, Neha,Mamta, Julie, etc.). Finally we would evaluate each alternative according to the

criteria and choose the one that best meets the criteria. We would thus have a schemelike this: decision whether ... select criteria ... identify alternatives ... make choice

However, the fact is that our decision whether to get married may really be acontingent decision. "I'll get married if I can find the right person." It will thus beinfluenced by the identification of alternatives, which we usually think of as a laterstep in the process.

Similarly, suppose we have arrived at the "identify alternatives" stage of the processwhen we discover that Neha (one of the girls identified as an alternative) has awonderful personality characteristic that we had not even thought of before, but thatwe now really want to have in a wife. We immediately add that characteristic to ourcriteria.

Thus, the decision making process continues to move back and forth, around andaround as it progresses in what will eventually be a linear direction but which in itsactual workings is highly recursive.


19/171

Decision-Making

Notes


Fig 3: Recursive Process of Decision Making

Simons ModelThe work of managers, of scientists, of engineers, of lawyers-the work that steers thecourse of society and its economic and governmental organizations-is largely work ofmaking decisions and solving problems. It is work of choosing issues that requireattention, setting goals, finding or designing suitable courses of action, and evaluatingand choosing among alternative actions. The first three of these activities-fixingagendas, setting goals, and designing actions-are usually called problem solving; thelast, evaluating and choosing, is usually called decision-making.

Any decision involves a choice selected from a number of alternatives, directedtoward an organizational goal or subgoal. Realistic options will have real


20/171


Notes


consequences consisting of personnel actions or non-actions modified byenvironmental facts and values.

In actual practice, some of the alternatives may be conscious or unconscious; some ofthe consequences may be unintended as well as intended; and some of the means andends may be imperfectly differentiated, incompletely related, or poorly detailed.

Over his lifetime, the academic scholar and Nobel price winner, Herbert Simon,explored the boundaries of human decision making in dynamic environments andcontributed significantly to management literature, economics, cognitive psychologyand artificial intelligence.

He questioned the assumptions behind the rational decision making process in whichindividuals clearly define the problem, generate and evaluate all alternative solutionsand select the best approach before implementing it.

He pointed out that people decide rationally only in a limited number of situations.They make choices according to their interpretation of the situation which is often asimplification. Rationality is "bounded", e.g. managers seldom have access to allrelevant information and must rely on a 'strategy of satisfying', that is to make the

best decision on limited information. They choose the first opportunity that seemssatisfactory rather than seek the best solution.Human rationality is limited:

1. Information is incomplete, imperfect or even misleading;

2. Problems are complex;

3. Human information processing is limited;

4. Time spent on decision making is limited;

5. Decision makers often have conflicting preferences for certain organizationalgoals.

In addition, organizations, themselves, act as "boundaries". Organizational boundaries allow organization members to focus and act without the need to (re)thinkthrough each action. Information in an organization is of two kinds:

1. Facts that can be verified with data;

2. Values that come from the mindsets embedded in the organizations culture andcommon approaches.

Simon saw humans as information processing entities wherein uncertainty comesfrom a lack of information. A fast changing and complex environment creates morecomplex problems that, in turn, require organisations and its members to expendgreater search efforts.

Depending on the circumstances, i.e. the complexity of the problem, the ambiguity ofthe decision making process and the number of decision makers, decision makersneed to adopt a more behavioural decision making process versus the more rational.

Simon suggests in his model (Fig 4) that decision-making process can be structuredinto three major phases:

1. The Intelligence Phase: wherein the decision maker looks for indications that aproblem exits.

2. The Design Phase: wherein the alternatives are formulated and analyzed.

3. Choice Phase: wherein one of the alternatives is selected and implemented.

Intelligence involves identifying the need for a decision or as Simon put it, searchingthe environment. During the intelligence phase information is gathered tounderstand the problem for which a decision is required, and the various


21/171

Decision-Making

Notes


assumptions that have to be made are made explicit. Once the environment has beensearched, i.e., the need for a decision identified, the design phase commences.

This comprises investigating and developing the problem domain and alternatives.During the design phase various alternatives are explored. Simons final phase is thatof choice, which describes the activity of selecting the most appropriate course ofaction from the alternatives previously generated. Finally in the choice phase, a bestor satisfactory decision is sought and selected, and some verification is undertaken.

The cycle of the stages is quite complex. Each phase in making a particular decision isitself a complex decision making process. For example, the design phase may call forfurther Intelligence, and any phase can generate new problems which themselveshave Intelligence, Design and Choice phases.

Fig 4: Simons Classic Model

Simons original work did not include implementation, which has been added later on(Fig. 5).

Fig 5: Simons Modified Model

Overall, Simon helps us to understand how the decisions of hundreds or thousands ofindividuals in an organization can be directed toward ultimate organizational goals.

Programmed versus Non-Programmed DecisionsManagers would be overwhelmed with decision making if each situation had to betreated as novel, that is, new. Fortunately this is not so. Managerial decisions can beprogrammed or non-programmed.


22/171


Notes


Programmed Decisions

Programmed decisions are made in routine, repetitive, well-structured situations withpredetermined decision rules. These may be based on habit, or established policies,rules and procedures and stem from prior experience or technical knowledge aboutwhat works or does not work in a given situation.

For example, organisations often have standardised routines for handling customercomplaints or employee discipline. Decisions are programmed to the extent that theyare repetitive and routine and that a definite approach has been worked out forhandling them. Because the problem is well-structured, the manager does not have togo to the trouble and expense of working through an involved decision makingprocess.

Non-programmed Decisions

Non-programmed decisions are unique decisions that require a 'custom made'solution. This is when a manager is confronted with an ill-structured or novelproblem and there is no 'cut and dried solution'. The creation of a marketing strategyfor a new service represents an example of a non-programmed decision. IBMAustralia's introduction of a personal computer in the 1980s was unlike any otherdecision the company had previously made.

Quantitative ToolsQuantitative techniques help a manager improve the overall quality of decision-making. These techniques are most commonly used in the rational/logical decisionmodel, but they can apply in any of the other models as well. Among the mostcommon techniques are decision trees, payback analysis, and simulations.

Decision trees

Decision tree is one of the most systematic tools of decision-making theory andpractice. Such trees are particularly helpful in situations of complex multistagedecision problems. For example, when you need to plan and organize a sequence ofdecisions and take into account how the choices made at earlier stages and theoutcomes of possible external events determine the types of decisions and events atlater stages of that sequence.

A decision tree shows a complete picture of a potential decision and allows a managerto graph alternative decision paths. Decision trees are a useful way to analyze hiring,marketing, investments, equipment purchases, pricing, and similar decisions thatinvolve a progression of smaller decisions. Generally, decision trees are used toevaluate decisions under conditions of risk.

The term decision tree comes from the graphic appearance of the technique that startswith the initial decision shown as the base. The various alternatives, based uponpossible future environmental conditions, and the payoffs associated with each of thedecisions branch from the trunk.

Decision trees force a manager to be explicit in analyzing conditions associated withfuture decisions and in determining the outcome of different alternatives. Thedecision tree is a flexible method. It can be used for many situations in whichemphasis can be placed on sequential decisions, the probability of various conditions,or the highlighting of alternatives.

A decision making tree is essentially a diagram that represents, in a speciallyorganized way, the decisions, the main external or other events that introduceuncertainty, as well as possible outcomes of all those decisions and events.


23/171

Decision-Making

Notes


Here is a schematic example that illustrates the basic elements of decision trees.

Fig 6: Decision Tree

Squares represent decisions you can make. The lines that come out of each square onits right show all the available distinct options that can be selected at that decisionanalysis point.

Circles show various circumstances that have uncertain outcomes (For example, sometypes of events that may affect you on a given path). The lines that come out of eachcircle denote possible outcomes of that uncontrollable circumstance. Write downabove each such line in the decision tree your best guesses for probabilities (forexample, 80% or 0.8) of those different outcomes.

Each path that can be followed along the decision tree, from left to right, leads tosome specific outcome. You need to describe those end results in terms of your maincriteria for judging the results of your decisions. Ideally, you will assign each endoutcome a quantitative measure of the overall total benefit you will receive from thatoutcome (you can express it as a perceived monetary value).

Now you have a complete decision making tree with specific numbers for both theprobabilities of the uncertain events and the benefit measures (desirability) of eachend result. At this stage the tree can give you more specific recommendation on whatwould be your best choices.

In particular, for each choice that you control (at the decision points shown bysquares), you can calculate the overall desirability of that choice. Just sum the benefitmeasures of all the end outcomes that can be traced back to that choice (via one path


24/171


Notes


or another), weighted by the probabilities of the corresponding paths. This will showyou the preferred choice (the one with the highest overall desirability).

If you have more than one decision point, you need to do that calculation for thedecisions that are at the latest stages first. Identify the choice that gives the highestoverall desirability and leave only that branch (removing the decision point). Do thesame with the remaining squares, working your way to the left (to the first decisionpoint in the sequence).

Payback Analysis

Payback analysis comes in handy if a manager needs to decide whether to purchase apiece of equipment. Say, for example, that a manager is purchasing cars for a rentalcar company. Although a less-expensive car may take less time to pay off, someclients may want more luxurious models.

To decide which cars to purchase, a manager should consider some factors, such asthe expected useful life of the car, its warranty and repair record, its cost of insurance,and, of course, the rental demand for the car.

Based on the information gathered, a manager can then rank alternatives based on thecost of each car. A higher-priced car may be more appropriate because of its longerlife and customer rental demand. The strategy, of course, is for the manager to choosethe alternative that has the quickest payback of the initial cost.

Many individuals use payback analysis when they decide whether they shouldcontinue their education. They determine how much courses will cost, how muchsalary they will earn as a result of each course completed and perhaps, degree earned,and how long it will take to recoup the investment. If the benefits outweigh the costs,the payback is worthwhile.

Simulations

Simulation is a broad term indicating any type of activity that attempts to imitate an

existing system or situation in a simplified manner. Simulation is basically model building, in which the simulator is trying to gain understanding by replicatingsomething and then manipulating it by adjusting the variables used to build themodel.

Simulations have great potential in decision making. In the basic decision-makingsteps, Step 4 is the evaluation of alternatives. If a manager could simulate alternativesand predict their outcomes at this point in the decision process, he or she wouldeliminate much of the guesswork from decision making.

Student Activity

1. What is PEST analysis?

2.

Which skills are required for good decision-making?3. What is decision-making tree?

SummaryDecision-making can be defined as the study of identifying and choosing alternatives

based on the values and preferences of the decision maker. Problem solving anddecision-making are important skills for business and life. Problem solving ofteninvolves decision-making, and decision-making is especially important formanagement and leadership.


25/171

Decision-Making

Notes


There are processes and techniques to improve decision-making and the quality ofdecisions. Decision-making is more natural to certain personalities, so the peopleshould focus more on improving the quality of their decisions. The six steps of thisnatural, intuitive decision-making process are:

Step 1: Define the problem

Step 2: Identify available alternative solutions to the problemStep 3: Evaluate the identified alternatives

Step 4: Make the decision

Step 5: Implement the decision

Step 6: Evaluate the decision

Programmed decisions are made in routine, repetitive, well-structured situations withpredetermined decision rules. Non-programmed decisions are unique decisions thatrequire a 'custom made' solution. Quantitative techniques help a manager improvethe overall quality of decision making. Among the most common techniques aredecision trees, payback analysis, and simulations.

KeywordsDecision-making: It can be defined as the study of identifying and choosingalternatives based on the values and preferences of the decision maker.

Decision-making process: The decision-making process is the process that is used tomake a decision.

Quantitative tools: These are the techniques which help a manager improve theoverall quality of decision making and these techniques are most commonly used inthe rational/logical decision model, but they can apply in any of the other models aswell.

Decision-making tree: It is essentially a diagram that represents, in a specially

organized way, the decisions, the main external or other events that introduceuncertainty, as well as possible outcomes of all those decisions and events.

Simulation: It is a broad term indicating any type of activity that attempts to imitatean existing system or situation in a simplified manner.

Review Questions1. What are the basic kinds of decisions?

2. What are the steps of decision-making process?

3. Write short notes on Simons Model.

4. Compare and contrast between Programmed versus Non-Programmed Decisions.

5. Describe decision tree with suitable diagram.6. What do you mean by quantitative tools? Discuss its importance.

7. What is payback analysis? Explain it.


26/171


27/171

Decision-Models

Notes


Unit 2 Decision-Models

Unit Structure Introduction Decision-Making Theory Decision-Making under Assumed Certainty Risk and Uncertainty of Decision Making Summary Keywords Review Questions Further Readings


Decision Making Theory Decision Making with assumed certainty Decision Making under uncertainty and risk Brief idea about risk

IntroductionDecisions are at the heart of success, and at times there are critical moments whenthey can be difficult, perplexing, and nerve racking. This unit provides help andguidance for making efficient and effective decisions by putting to use a well-structured approach and well-focused process known as the modeling or paradigmprocess.

The word paradigm comes from the Greek word paradeigma, meaning "model" or"pattern." A model represents a way of looking at the world, a shared set ofassumptions that enable us to understand or predict behavior.

Models have a powerful influence on individuals and on society because our view ofthe world is determined by our set of assumptions about it. To put it another way, ourvision is often affected by what we believe about the world; our beliefs oftendetermine the information that we "see."

Decision-Making TheoryA broad range of concepts which have been developed to both describe and prescribethe process of decision making, where a choice is made from a finite set of possiblealternatives. Normative decision theory describes how decisions should be made inorder to accommodate a set of axioms believed to be desirable; descriptive decisiontheory deals with how people actually make decisions; and prescriptive decisiontheory formulates how decisions should be made in realistic settings.


28/171


Notes


Thus, this field of study involves people from various disciplines: behavioral andsocial scientists and psychologists who generally attempt to discover elaboratedescriptive models of the decision process of real humans in real settings;mathematicians and economists who are concerned with the axiomatic or normativetheory of decisions; and engineers and managers who may be concerned withsophisticated prescriptive decision-making procedures.

Most of decision theory is normative or prescriptive, i.e. it is concerned withidentifying the best decision to take, assuming an ideal decision maker who is fullyinformed, able to compute with perfect accuracy, and fully rational.

The practical application of this prescriptive approach (how people should makedecisions) is called decision analysis, and aimed at finding tools, methodologies andsoftware to help people make better decisions. The most systematic andcomprehensive software tools developed in this way are called decision supportsystems.

Since it is obvious that people do not typically behave in optimal ways, there is also arelated area of study, which is a positive or descriptive discipline, attempting todescribe what people will actually do. Since the normative, optimal decision often

creates hypotheses for testing against actual behaviour, the two fields are closelylinked.

Furthermore it is possible to relax the assumptions of perfect information, rationalityand so forth in various ways, and produce a series of different prescriptions orpredictions about behaviour, allowing for further tests of the kind of decision-makingthat occurs in practice.

The conditions for making decisions can be divided into two types, certainty anduncertainty. Decisions made under certainty or uncertainty is based on our feelingsand our experiences.

Decision-Making under Assumed Certainty

Virtually all decisions are made in an environment to at least some uncertainty.However, the degree will vary from relative certainty to great uncertainty. There arecertain risks involved in making decisions.

In a situation involving certainty, people are reasonably sure about what will happenwhen they make a decision. The information is available and is considered to hereliable, and the cause and effect relationships are known.

We experience certainty about a specific question when we have a feeling of complete belief or complete confidence in a single answer to the question. Decisions such asdeciding on a new carpet for the office or installing a new piece of equipment orpromoting an employee to a supervisory position are made with a high level ofcertainty. There is always some degree of uncertainty.

The term decision-making under certainty can be used in a situation when for eachdecision alternative there is only one event and therefore only one outcome for eachaction.

Risk and Uncertainty of Decision-MakingA decision under uncertainty is when there are many unknowns and no possibility ofknowing what could occur in the future to alter the outcome of a decision. We feeluncertainty about a situation when we can't predict with complete confidence whatthe outcomes of our actions will be. We experience uncertainty about a specificquestion when we can't give a single answer with complete confidence.


29/171

Decision-Models

Notes


Launching a new product, a major change in marketing strategy or opening your first branch could be influenced by such factors as the reaction of competitors, newcompetitors, technological changes, changes in customer demand, economic shifts,government legislation and a host of conditions beyond your control.

These are the type of decisions facing the senior executives of large corporations whomust commit huge resources. In a situation of uncertainty people have only a meagerdatabase, they do not know whether or not the data are reliable, and they are veryunsure about whether or not the situation may change.

Moreover, they cannot evaluate the interactions of the different variables. Forexample, a corporation that decides to expand its Operation to an unfamiliar countrymay know little about the countrys culture, laws, economic environment, andpolitics.

Making decisions involves a degree of risk, it would be helpful to examine risk andrisk analysis in this chapter in order to gain an understanding of what is involved.Risk and uncertainty create anxiety, yet they are necessary components of an activelife.

RiskMost of us hate the idea of risk. While we collectively spend a great deal of time andmoney to reduce it, we can never hope to eliminate it. The reason is that some amountof uncertainty is "built in" to all aspects of the world around us. In fact, at the smallestlevel of physical reality, quantum physicists must deal only in probabilities, as theycannot predict with certainty which events will occur, or where or when they mightoccur.

The very word "risk" tends to make us nervous, as it portends the probability thatsomething bad may happen to us. The word "happen" is also a clue to our deepconcerns about risk, because it indicates events that are out of our control.

When asking the managers how they defined risk, most of them distinguished between different types of risks, such as fire risk, financial risk, technical risk,commercial risk, and investment risk. They said that a risky situation is a situationwhere the outcome is unknown to the decision-maker, i.e. he/she is not sure whichoutcome will occur and the uncertainty leads to erroneous choices.

When the managers were asked to describe a risky decision they had recently made,or a risky situation they had been involved in, more than half of them associated thiswith different kinds of investment activities and divided them into such categories as:

Investing in new machines and techniques Acquisition of new companies Development of new products and entering new markets.

They were uncertain about whether they would reach the expected production speed

within the scheduled time, if they would be able to produce top quality paper, and thereliability of the new machines. One manager said, New techniques are alwaysriskier than old techniques.

So, we must decide if we, for example, want to be first in a new market or the firstwith a new product, or if we should hold back for a while and enter the market asnumber two. Another risky area pointed out by a manager was that they were veryvulnerable concerning issues related to information technology.

One problem that a manager did bring up is related to the acquisition of othercompanies. He said, I do not think that we really are aware of how to estimatedifferent types of risk that we need to deal with.


30/171


Notes


He also said that even though the mathematical part of many problems was easilysolved since they have figures concerning the cash flow, the potential developmentand so on, they are still greatly governed by the soft aspects of the decision-makingprocess. He also said that they often invest in projects that they believe will be goodinvestments, and that they do not only focus on figures or the investment index.Three others expressed the same sentiments concerning the acquisition of newcompanies by saying that they sometimes even ignore the figures they have and basetheir decisions on their gut reaction.

One example of risk, which is difficult to estimate and predict, is when to leave anexisting market. The risky element in such a case is that once you have left a marketyou can not return. One manager, who refers to such a case concerning entering anew market with newspaper-paper, said, These kinds of decisions are veryunreliable. Therefore, many decisions of that type are based on subjective appraisalsof the decision-makers not on any calculations. Regarding the future interest ratesrisk one of them said, We used to consult a bank and some other institutionsregarding these kinds of matters, but we make the final decision by gut feelings, i.e.,we choose the alternative that feels good.

Student Activity1. What are the conditions for making decisions?

2. What are the classifications of the conditions for making decisions?

3. Name three different types of risk.

SummaryNormative decision theory describes how decisions should be made in order toaccommodate a set of axioms believed to be desirable; descriptive decision theorydeals with how people actually make decisions; and prescriptive decision theoryformulates how decisions should be made in realistic settings. The practicalapplication of this prescriptive approach (how people should make decisions) iscalled decision analysis, and aimed at finding tools, methodologies and software tohelp people make better decisions. The conditions for making decisions can bedivided into two types, certainty and uncertainty. Decisions made under certainty oruncertainty is based on our feelings and our experiences. There are certain risksinvolved in making decisions. When asking the managers how they defined risk, mostof them distinguished between different types of risks, such as fire risk, financial risk,technical risk, commercial risk, and investment risk.

Keywords Model: It represents a way of looking at the world, a shared set of assumptions that

enable us to understand or predict behavior. Normative decision theory: This theory describes how decisions should be made inorder to accommodate a set of axioms believed to be desirable; descriptive decisiontheory deals with how people actually make decisions; and prescriptive decisiontheory formulates how decisions should be made in realistic settings.


31/171

Decision-Models

Notes


Review Questions1. Most of decision theory is normative or prescriptive justify the statement.

2. What is risk? Describe the meaning of risky situation in your own word.

3. Discuss decision making under uncertainty.

Further ReadingsDhiraj Sharma Foundation of IT Published by Excel Books

Deepak Bharioke Foundation Information Technology Published by Excel Books

Ralph Kimball and Margy Ross The Data Warehouse Toolkit: The Complete Guide to Dimensional Modeling Published by Wiley

Larissa T. Moss and Shaku Atre Business Intelligence Roadmap: The Complete Project Lifecycle forDecision-Support Applications - Published by Addison-Wesley


32/171

Introduction to DSS

Notes


Unit 3 Introduction to

DSSUnit Structure

Introduction Concept of DSS Characteristics and Objectives of DSS Advantage and Limitation of Decision Support System Comparison with EDP/MIS

Summary Keywords Review Questions Further Readings


Concept of DSS Characteristic and objectives of DSS Advantages and limitation of DSS How DSS differs with EDP and MIS

IntroductionInformation age has revolutionized the way companies think about and serve theircustomers. Companies leverage technology to provide a wide array of services, fromautomated customer support to virtual stores on the Internet. In the process,

businesses have been collecting data at customer-transaction level. With the advent ofsophisticated data mining tools, companies have started using data to identify, attract,and retain profitable customers.

Decision support systems are powerful tools integrating scientific methods forsupporting complex decisions with techniques developed in information science, andare gaining an increased popularity in many domains.

They are especially valuable in situations in which the amount of availableinformation is prohibitive for the intuition of an unaided human decision maker andin which precision and optimality are of importance. Decision support systems aidhuman cognitive deficiencies by integrating various sources of information, providingintelligent access to relevant knowledge, aiding the process of structuring, andoptimizing decisions.

Concept of DSSDecision support systems integrate technology and business process to deliverconsistent, reliable, and efficient decision-making process. At the heart of all


33/171


Notes


sophisticated decision support systems lie complex statistical and mathematicalmodels.

These models capture historical trends, incorporate predictive elements, and quantifythe impact of strategic and tactical business decisions in terms of market share andprofitability. The degree of sophistication can vary based on the purpose of thesystem. A sample process flow for a full-fledged decision support system, completewith forecasting and optimization models, is shown on below.

Fig 1: Overview of DSS

Making decisions concerning complex systems (e.g., the management oforganizational operations, industrial processes, or investment portfolios; thecommand and control of military units; the control of nuclear power plants) oftenstrains our cognitive capabilities. Even though individual interactions among asystem's variables may be well understood, predicting how the system will react to anexternal manipulation such as a policy decision is often difficult.

What will be, for example, the effect of introducing the third shift on a factory floor?One might expect that this will increase the plant's output by roughly 50%. Factorssuch as additional wages, machine weardown, maintenance breaks, raw materialusage, supply logistics, and future demand also need to be considered, however,

because they will all affect the total financial outcome of this decision. Many variablesare involved in complex and often subtle interdependencies, and predicting the totaloutcome may be daunting.


34/171

Introduction to DSS

Notes


There is a substantial amount of empirical evidence that human intuitive judgmentand decision making can be far from optimal, and it deteriorates even further withcomplexity and stress. In many situations, the quality of decisions is important;therefore, aiding the deficiencies of human judgment and decision making has been amajor focus of science throughout history.

Disciplines such as statistics, economics, and operations research developed variousmethods for making rational choices. More recently, these methods, often enhanced

by various techniques originating from information science, cognitive psychology,and artificial intelligence, have been implemented in the form of computer programs,either as stand-alone tools or as integrated computing environments for complexdecision making. Such environments are often given the common name of decisionsupport systems (DSSs).

The concept of DSS is extremely broad, and its definitions vary, depending on theauthor's point of view. To avoid exclusion of any of the existing types of DSSs, wedefine them roughly as interactive computer-based systems that aid users in

judgment and choice activities. Another name sometimes used as a synonym for DSSis knowledge-based systems, which refers to their attempt to formalize domainknowledge so that it is amenable to mechanized reasoning.

Decision support systems are gaining an increased popularity in various domains,including business, engineering, the military, and medicine. They are especiallyvaluable in situations in which the amount of available information is prohibitive forthe intuition of an unaided human decision maker, and in which precision andoptimality are of importance.

Decision support systems can aid human cognitive deficiencies by integrating varioussources of information, providing intelligent access to relevant knowledge, and aidingthe process of structuring decisions. They can also support choice among well-definedalternatives and build on formal approaches, such as the methods of engineeringeconomics, operations research, statistics, and decision theory. They can also employartificial intelligence methods to heuristically address problems that are intractable byformal techniques.

Proper application of decision-making tools increases productivity, efficiency, andeffectiveness, and gives many businesses a comparative advantage over theircompetitors, allowing them to make optimal choices for technological processes andtheir parameters, planning business operations, logistics, or investments. Although itis difficult to overestimate the importance of various computer-based tools that arerelevant to decision making (e.g., databases, planning software, spreadsheets).

There is a substantial amount of empirical evidence that human intuitive judgmentand decision making can be far from optimal, and it deteriorates even further withcomplexity and stress. Because in many situations the quality of decisions isimportant, aiding the deficiencies of human judgment and decision making has beena major focus of science throughout history.

Disciplines such as statistics, economics, and operations research developed variousmethods for making rational choices. More recently, these methods, often enhanced

by a variety of techniques originating from information science, cognitive psychology,and artificial intelligence, have been implemented in the form of computer programs,either as stand-alone tools or as integrated computing environments for complexdecision making. Such environments are often given the common name of decisionsupport systems (DSSs).

The concept of DSS is extremely broad, and its definitions vary, depending on theauthor's point of view. To avoid exclusion of any of the existing types of DSSs, we willdefine them roughly as interactive computer-based systems that aid users in

judgment and choice activities. Another name sometimes used as a synonym for DSS


35/171


Notes


is knowledge-based systems, which refers to their attempt to formalize domainknowledge so that it is amenable to mechanized reasoning.

Decision support systems are interactive, computer-based systems that aid users in judgment and choice activities. They provide data storage and retrieval but enhancethe traditional information access and retrieval functions with support for model

building and model-based reasoning. They support framing, modeling, and problemsolving.

Typical application areas of DSSs are management and planning in business, healthcare, the military, and any area in which management will encounter complexdecision situations. Decision support systems are typically used for strategic andtactical decisions faced by upper-level management - decisions with a reasonably lowfrequency and high potential consequences in which the time taken for thinkingthrough and modeling the problem pays off generously in the long run.

Characteristics and Objectives of DSSThe objective of DSS is to involve the manager/decision maker in the decision-analysis process while simultaneously relieving that person of the burden ofdeveloping and performing detailed analysis. DSS represents a convergence of thetechnologies and bodies of knowledge in the following areas:

Computer hardware and software technology, especially microprocessor systems Data processing and information systems theory and applications Management-science or operations-research modeling and analysis techniques,

including both analytical- and simulation-modeling approaches.

Finally, goals that are included in the DSS (aggregate, individual, and comparison)are identified along with factors considered in goal formulations.

Characteristics of a DSS include the following: It is computer based It is interactive It includes a user friendly command language It utilizes models There is easy access to databases It can use graphics It allows a flexible decision-analysis process It supports managerial judgment. It supports for decision makers in semi-structured and unstructured problems. It supports managers at all levels. It supports individuals and groups. It supports for interdependent or sequential decisions. It supports intelligence, design, choice, and implementation. It supports variety of decision processes and styles. DSS should be adaptable and flexible. DSS should be interactive and provide ease of use. Effectiveness balanced with efficiency (benefit must exceed cost). Complete control by decision-makers.


36/171

Introduction to DSS

Notes


Ease of development by (modification to suit needs and changing environment)end users.

It supports modeling and analysis. Data access. Standalone, integration and Web-based.

Advantage and Limitation of Decision SupportSystem

Auditing Advantage: Decision support system delivers added advantage in terms ofresults tracking and auditing. Senior management can easily track the history ofchanges and the impact of the changes on market share and profits thorough plethoraof reports built into the systems. Also, all changes to the systems can be easily storedin databases for future audits.

Reliability Advantage: Evidently, the underlying analytical models are the keydrivers of delivering reliable results. Carefully calibrated models capture key business

drivers and help managers arrive at decisions identifying right trade-off the meetstrategic business objectives. For example, while developing pricing strategy for aregion, one can expect similar results from any pricing manager as long as the

business rules in the system are set identical

Consistency Advantage: Since the underlying modeling assumptions and businessrules are set at the beginning of the decision process, each scenario will produceconsistent results. Most of the business rules are input as option though graphicaluser interface and the range of the input variables are set using strict statistical and

business criteria. Models are built to alert users when inconsistent data is used asinput. Thus, by minimizing user input errors and automating model parameters,these systems deliver consistent results scenario after scenario.

Efficiency Advantage: Scenario analyses are critical to not only quantify the impact of

variety of business decisions, but also to get a handle around the sensitivity of themodel results. In todays fast paced environment, it is speed is a critical element ofcompetitive response. Decision support systems deliver efficiency by giving managersconduct scenario analyses very effectively without having to compromise on thequality of the results. In other words, managers can focus on strategy and results, notdata input.

Limitations: Decision support systems work the best where business processes arewell defined and are followed in a systematic fashion. Inconsistent processes aredifficult to model and automate. Second requirement is availability of reliable andstatistically significant data. Since the model calibration is highly sensitive to thequality and quantity of the data used, it is important to have best data set available formodeling purpose. Businesses will have to hire high caliber employees who can

handle the sophistication of decision support systems, not only to run complexscenarios, but also analyze and interpret the results. Last but not the least, thesesystems are as good as the models that drive the results; hence periodic recalibrationof model parameters is a highly recommended.

Comparison with EDP/MISElectronic Data Processing (EDP) is the first of several links of the data-to-informationchain. With EDP we merely access, store, retrieve, and manipulate data, a functionwell suited to clerical-level and operational-level endeavors. Before computers, wedid the data processing function well using hand-generated spreadsheets for


37/171


Notes


manipulation and notebooks or file cabinets for storage. Computers are able toduplicate this function faster and more consistently.

One disadvantage is that since we now can do data processing so much faster, we cando so much more of it. So we produce the same data, manipulated into dozens orhundreds of different spreadsheets or tables.

Therefore, word processors are computer-based EDP devices. When you think aboutit, state-of-theart networking, spreadsheeting, and windowing practices are nothingmore than manipulating data, since moving data from place to place is a form ofmanipulation. EDP is important to consider because that is most of what we are reallydoing under the guise of MIS or DSS.

MIS Is More Than Manipulating Data. When we look at the technological advanceswe are so excited about today, we see us manipulating data, not making information.Networking is moving data from place to place, windowing is displaying data, andspreadsheeting is aggregating data. Either by moving it, showing it, or tallying it, weare getting better and better at just manipulating raw data, not enhancing its value ortruly supporting management.

The Management Information System (MIS) was coined to represent a more-useful,higher level form of management support using information rather than just data.Unfortunately, when MIS was required to do much more than EDP, MIS failed. Thatis, MIS may have been a new term, but it did little else than raise managersexpectations and sell a lot of computers. (Those of us in the information business willpay for that). MIS is the entire data-to-information chain and includes not only theEDP links, but the links for forming and presenting information. MIS never hasadequately addressed the measurement/data and the information portrayal/information perception interfaces. In terms of the individual links in the chain, wevedeveloped the hardware and software specialties far beyond the ability or need ofmost of us to fully use them.

Decision Support Systems (DSS) arose as a response to bad feelings about MIS. Giventhat MIS hasn't progressed much from EDP, why should DSS suddenly be able to

accomplish more than what was originally expected of MIS? What is the difference between information for managers and support for decisions in the real world wherewe are doing neither?

DSS is quite different from MIS and includes all the tools of the what is used tomanage component; and DSS does address the measurement-to-data and theinformation-portrayal-to-information-perception interfaces.

The key to DSS is the synergism that results from the tools working well together.Thus, DSS focuses on the interrelatedness of the tools. The methods category ofmanagement tools should affect the plans in the guides and rules category. The plansshould be used as much as the data-to-information chain because these two toolsshould be used hand-in-glove in formulating the reference points so we can executeagainst them using our MIS.

Often, in an effort to obtain computerized decision support, managers gain such agood understanding of these interactions, the need for computerization (orautomation) is lessened because the manager has systematized what he uses tomanage.


38/171

Introduction to DSS

Notes


Fig 2: Relationship between EDP/MIS/DSS

In short, the DSS supports who manages, or the decision maker, and MSS supportswhat is managed, or the operation. A comparison between EDP, MIS and DSS is given

below in the tabular form:

Parameters EDP MIS DSSTarget Users Basic level

operatorsMiddle managers Top managers and

executivesFocus Data, files,

storage,transactionprocessing, andreports

Information retrieval, Planand analyze data againstexpected values, Integration

What if analysisthrough use of models,generation of decisionalternatives

Characteristics Basic personnelinformation

Inquiry capability, report-generation capability

Interactive for users

Examples Payroll Turnover reports, age andgender distribution, EEOcompliance report

Human resourceplanning,compensationsimulation

Human resourceplanning

Skills inventory Turnover analysis,Organizational charting

Succession planning,Work force dynamicsanalysis

Staffing Basic employeeinformation,

Applicant tracking

Recruitment analysis,Selection analysis, Positionanalysis, Manpowerstructure analysis

Staffing simulation

Training andcareerdevelopment

Employee trainingdata, Trainingcourses Careerprofile

Training needs analysis,Training cost-benefitanalysis, Promotion analysis

Career managementsimulation, Trainingevaluation anddecisions

Contd

Executives

Analysis, experts

Functional middle managers

Employees charged with operativeresponsibilities

EIS

DSS DSS DSS

MIS MIS MIS

EDP/TPS EDP/TPS EDP/TPS


39/171


Notes


Performancemanagement

Performance data Performance appraisalanalysis, Attitude survey,

Attendance managementanalysis, Productivityanalysis

Performancemanagementsimulation

Compensation

management

Payroll, Health

insurance Routinereports (e.g.income tax)

Personnel cost analysis

Compensation structureanalysis

Compensation

managementsimulation

Student Activity

1. What is DSS?

2. What is knowledge-based system?

3. What is the goal of DSS?

4. What is EDP?

5. What is MIS?

SummaryIn the process, businesses have been collecting data at customer-transaction level.Decision support systems are powerful tools integrating scientific methods forsupporting complex decisions with techniques developed in information science, andare gaining an increased popularity in many domains. Decision support systems aidhuman cognitive deficiencies by integrating various sources of information, providingintelligent access to relevant knowledge, aiding the process of structuring, andoptimizing decisions. Decision support systems integrate technology and businessprocess to deliver consistent, reliable, and efficient decision-making process. At theheart of all sophisticated decision support systems lie complex statistical andmathematical models. Decision support systems are gaining an increased popularityin various domains, including business, engineering, the military, and medicine.Decision support systems can aid human cognitive deficiencies by integrating varioussources of information, providing intelligent access to relevant knowledge, and aidingthe process of structuring decisions. Decision support systems are interactive,computer-based systems that aid users in judgment and choice activities. Theyprovide data storage and retrieval but enhance the traditional information access andretrieval functions with support for model building and model-based reasoning. Theysupport framing, modeling, and problem solving.

KeywordsDecision Support Systems: Those are powerful tools integrating scientific methods forsupporting complex decisions with techniques developed in information science, and

are gaining an increased popularity in many domains. Knowledge-based systems: It refers to the systems which attempt to formalize domainknowledge so that it is amenable to mechanized reasoning.

Electronic Data Processing (EDP): It is the first of several links of the data-to-information chain by which we merely access, store, retrieve, and manipulate data.

Management Information System (MIS): It was coined to represent a more-useful,higher level form of management support using information rather than just data.


40/171

Introduction to DSS

Notes


Review Questions1. What are the characteristic features of DSS?

2. Discuss the advantages and disadvantages of DSS.

3. Compare and contrast between EDP, MIS and DSS.

4. Proper application of decision-making tools increases productivity, efficiency,and effectiveness, and gives many businesses a comparative advantage justifythe statement.

Further ReadingsDhiraj Sharma Foundation of IT Published by Excel Books

Deepak Bharioke Foundation Information Technology Published by Excel Books

Ralph Kimball and Margy Ross The Data Warehouse Toolkit: The Complete Guide to Dimensional Modeling Published by Wiley

Larissa T. Moss and Shaku Atre Business Intelligence Roadmap: The Complete Project Lifecycle forDecision-Support Applications - Published by Addison-Wesley


41/171

Levels of Decision

Support System

Notes


Unit 4 Levels of

DecisionSupport System

Unit Structure Introduction Levels of Decision Support System Forecasting Packages Statistical Packages Relationship Summary Keywords Review Questions Further Readings


Overview of different levels of DSS Concept of different DSS related software packages

IntroductionIn order to provide efficient adjustment of the business system to the changes in themarket, one must pay special attention to the processes of solving complexmanagement problems. Considering that there are established procedures and waysof solving simple problems in business practice, it is also important to establish aprocedure for solving complex management problems.

The solving of complex management problems conducted by established procedure,i.e. by defined phases, makes the whole act a lot more easier for the decision maker, because it directs him how to organize the problem solving activities, shortens the

time of problem solving, increases the quality of decisions made, whereby theunwanted results are less likely to happen.

The unit describes the framework of decision support systems to solving complexmanagement problems in business systems. The aim of decision support systemapplication is to help the decision makers in the process of solving complexmanagement problems and to improve the quality of decisions made in suchcircumstances.


42/171


Notes


Levels of Decision Support SystemDecision Support Systems should be defined as a broad category of informationsystems for informing and supporting decision-makers. DSS are intended to improveand speed-up the processes by which people make and communicate decisions.

We need to improve how we define Decision Support Systems on both a conceptuallevel and on a concrete, technical level. Both managers and DSS designers need tounderstand categories of decision support so they can better communicate about whatneeds to be accomplished in informing and supporting decision makers.

DSS systems are not entirely different from other systems and require a structuredapproach. A framework was provided by Sprague and Watson (1993). The frameworkhas three main levels:

1. Technology levels

2. People involved

3. Developmental approach

Technology LevelsSprague has suggested that there are three levels of hardware and software that has been proposed for DSS.

Specific DSS

This is the actual DSS application that a user (decision maker) interacts with to solve aspecific decision problem. This is the part of the application that allows the decisionmaker to make decisions in a particular problem area. The user can act upon thatparticular problem.

DSS Generator

This level contains Hardware/software environment that allows people to easilydevelop specific DSS applications. It is an integrated development software packagethat provides a set of capabilities to build a specific DSS quickly, inexpensively, andeasily. Examples are IFPS/Plus, Encore!, Nomad, Excel, Lotus-123, and Access. Thislevel makes use of case tools or systems such as Crystal, AIMMS, iThink andClementine.

DSS Tools

This is the lowest level of DSS technology. The underlying technical building blocks(graphics packages, data base management systems, and so on) for both thegenerators and applications (Sprague and Watson [1989]).

These tools are used to develop DSS Generators and/or Specific DSS applications.Examples are programming languages (COBOL, C++, J++, Visual Basic, etc.),operating systems (UNIX, Windows 95 or 98, NT, LINUX, etc.), and networkoperating systems (NOS). This level contains lower level hardware/software. DSSgenerators including special languages, function libraries and linking modules.


43/171

Levels of Decision

Support System

Notes


Fig 1: Technical level of DSS

People Involved

Sprague suggests there are five roles involved in a typical DSS development cycle:

1.

The end user.2. An intermediary.

3. DSS developer

4. Technical supporter

5. Systems Expert

Developmental Approach

The developmental approach for a DSS system should be strongly iterative. This willallow for the application to be changed and redesigned at various intervals. T

MBA 544-Advanced Decision Support System

Documents