QADM - Decision Making

Embed Size (px)

Citation preview

  • 8/8/2019 QADM - Decision Making

    1/19

    DDECISIONECISION AANALYSNALYSIISSGroup MembersAbdul Mulla

    Dipti BharwadaPrachi Jadhav

    Vaibhav Patki

    Vinit Singh

  • 8/8/2019 QADM - Decision Making

    2/19

    A good decision is one that is based on logic, considers all available

    data and possible alternatives, and applies the qualitative and

    quantitative approaches to solve them

    Decision Analysis (DA) provides structure and guidance for thinking

    systematically about hard decisions

    It allows a decision maker to take action with confidence gainedthrough a clear understanding of the problem

    Decision Analysis helps to improve the quality of the resulting

    decisions

  • 8/8/2019 QADM - Decision Making

    3/19

    The Five commonly used decision criteria are as follows:

    Maximin Criterion

    Maximax Criterion

    Hurwiczs alpha Criterion

    Savages minimax regret Criterion

    Laplaces Criterion

  • 8/8/2019 QADM - Decision Making

    4/19

    Decision making under Certainty or Deterministic situation

    Decision making under Risk or Stochastic situation

    Decision making under Uncertainty

    Conflict and Competitive situation (Games)

  • 8/8/2019 QADM - Decision Making

    5/19

    Decision Making under Certainty

    In this case the decision maker tends to maximize returns or

    minimize costs. The decision taken is such that it satisfies such

    criteria in a problem where he knows the outcomes with

    certainty.

    Decision Making under Uncertainty

    In this case the problem is uncertain i.e. , the decision maker

    cannot estimate or anticipate the probability of occurrence of the

    events with each decision alternatives.

  • 8/8/2019 QADM - Decision Making

    6/19

    Decision Making under Risk

    In this case the decision maker can anticipate theprobability of occurrence of events that he cannot

    control which is called State of Nature associated with

    each decision alternative.

  • 8/8/2019 QADM - Decision Making

    7/19

    y A decision problem is characterized by decision alternatives, states

    of nature, and resulting payoffs.

    y The decision alternatives are the different possible strategies the

    decision maker can employ.

    y The states of nature refer to future events, not under the control of

    the decision maker, which may occur. States of nature should bedefined so that they are mutually exclusive and collectively

    exhaustive.

  • 8/8/2019 QADM - Decision Making

    8/19

    The consequence resulting from a specific combination of a

    decision alternative and a state of nature is a payoff.

    A table showing payoffs for all combinations of decision alternatives

    and states of nature is a payoff table.

    Payoffs can be expressed in terms of profit, cost, time, distance orany other appropriate measure.

  • 8/8/2019 QADM - Decision Making

    9/19

    In the fall farmer is offered Rs. 50000 for his Apple crop, which will

    be harvested in the beginning of the following year. If the farmer

    accepts the offer, the money is his regardless of the quality orquantity of the harvest. If the farmer does not accept the offer, he

    must sell his oranges in the open market after they are harvested.

    Under normal conditions, the farmer can anticipate receiving Rs.

    70000 on the open market for his harvest. If he experiences a Frost,

    however then much of his harvest will be ruined & he can anticipate

    receiving only Rs. 15000 in the open market.

  • 8/8/2019 QADM - Decision Making

    10/19

    Determine recommended decisions, using each of the criterion

    under the situation of uncertainty. ( for Hurwicz- alpha criterion, use

    alpha= 0.4)

    Determine recommended decision under the appropriate criterion if

    in the past, the farmer has lost much of his harvest to frost 1 out of

    every 7 years.

  • 8/8/2019 QADM - Decision Making

    11/19

    Action Frost No Frost

    Accept 50,000 50,000

    Do Not Accept 15,000 70,000

  • 8/8/2019 QADM - Decision Making

    12/19

    Action Minimum Payoff

    Accept 50,000

    Does N

    ot Accept

    15,000

    According to Maximin criterion decision is to Acceptthe offer.

    Maximin Value

  • 8/8/2019 QADM - Decision Making

    13/19

    Action Maximum Payoff

    Accept 50000

    Does NotAcc

    ept 70000 Maximax Value

    According to Maximax Criterion recommended decision isto Not Accept the offer.

  • 8/8/2019 QADM - Decision Making

    14/19

    Action Expected Payoff

    Accept 50000

    Does not Accept 42500

    According to Laplace Criterion the recommendeddecision is toAccept the offer.

  • 8/8/2019 QADM - Decision Making

    15/19

    Frost No Frost Expected

    Payoff

    Accept 50000 50000 50000

    Does not

    Accept

    15000 70000 62143

    Probability 1/7 6/7

    According to EMV criterion the recommended

    decision is do not accept the offer.

    The corresponding EP is Rs. 62143.

  • 8/8/2019 QADM - Decision Making

    16/19

    Action Frost No Frost

    Accept 0 20000

    Do Not Accept 35000 0

    Action MaximumRegret

    Accept 20000

    Do Not Accept 35000

    According to the Savage Criterion

    recommended decision is to Accept the offer.

    Minimax

    Regret

  • 8/8/2019 QADM - Decision Making

    17/19

    y Let = 0.4

    Action Weighted payoff

    Accept 50000*0.4 + 50000*0.6 = 50000

    Do notAccept 70000* 0.4 + 15000* 0.6 = 37000

    According to Hurwicz- alpha Criterion when is 0.4,the recommendation is toAccept the offer.

  • 8/8/2019 QADM - Decision Making

    18/19

    y A Decision Tree is a chronological representation of the decision

    problem.

    y Each decision tree has two types of nodes; Round nodescorrespond to the states of nature while Square nodes correspond

    to the decision alternatives.

    y The Branches leaving each round node represent the different

    states of nature while the branches leaving each square noderepresent the different decision alternatives.

    y At the end of each limb of a tree are the payoffs attained from the

    series of branches making up that limb. Squares orRectangles

    depict decision nodes.

  • 8/8/2019 QADM - Decision Making

    19/19