Decision and Risk Analysis (Group 3 & 11) [PDF]

Citation preview

SAPPK



Decision And Risk Analysis



ELISABETH KIRANA 25422037 ANNISYA ALIFVIA 25422020



MUHAMMAD DAFFA 25422003



Group 03



ALI AMMARULLAH 25422008



ANDHYKA PUTRA ARIE GAYO 25422042



ROIS DINAN 25422025



Group 11



Outline



Decision Analysis A group of methods or instruments for analysis that help with decisionmaking and help choose the best course of action for allocating limited resources in the most efficient manner. A methodical method of problem-solving with the least amount of negative impact involves selecting the best option from a number of available options. The chosen option is then used by the decision-maker.



Decision and Optimization Decisions involve deciding on the course of action to be taken, and the ideal decision cannot be determined until its implementation has produced consequences. Given our preferences and available knowledge, it is vital to select the option that is the most logical or optimal selection. Optimization is a decision-making criterion that produces the optimal outcomes from the available options or conditions. The definition of optimal is "meeting requirements/achieving expected target criteria."



Decision Making Process Prior to selecting a choice, the decision-maker might consider several options thanks to the decision-making process. The process is as follows: The issue is identified, and all workable solutions are taken into account. Each alternative's potential results are assessed. Discussions of outcomes center on their financial benefits (payoffs) or net gains in terms of resources or time. Different uncertainties are measured using probability. The accuracy of the judgements determines how effective the ideal plan will be. The decision-maker should recognize and assess the optimal strategy's sensitivity to the critical variables.



Classification Criteria Decision Making



Decision Making Under Certainty Future events are known. The decision maker knows the consequences of each alternative. Decision Making Under Risk There is a little knowledge about the possibility of the event (state of nature) that will occur. There is a probability but it is not known which event will occur. Decision Making Under Uncertainty There is no information or knowledge about the possibility that will occur (natural conditions) in the future.



Risk Risk indicates a certain amount of uncertainty and the inability to completely control the results or effects of a certain action. In a risky situation, the decision-maker has limited knowledge of the options that are accessible, but he or she has an understanding of the likelihood of the results for each option based on the information that is currently available or from prior experience. In risk-based decision making, risk estimates and risk narratives form the basis for decisions. Unacceptable levels of risk trigger action. By contrast, risk-informed decision-making trades off levels of risk with other criteria to arrive at a decision



Risk By correctly managing risk, one can only reduce it as far as one can within



specific bounds without totally eliminating it. However, in some instances



the elimination of one risk may increase some other risks. The assessment of a risk and its subsequent impact on the decision-



making process are necessary for effective risk management.



Type of Decision Making in Uncertain and Risk Condition Risk Averse Maintain safe conditions to avoid high risk of loss Risk Taking Dare to take a chance with high risk Risk Neutral Name Job Title



Accept the risk but do not take the risk to get a higher return



Making Decision Under Uncertainty and Risk When making a decision under uncertain conditions, the probability of occurrence of a possibility is also uncertain. when these probabilities cannot be estimated, decision choices that depend on these inestimable probabilities are called decision-making under conditions of risk.



There are five keys concept that becomes the foundation of Decision Analysis : Decision Alternatives Decision analysis is the action we can take. for example, to invest on something or not, to bet or something and not. There will be minimum of 2 decision alternatives, but no maximum limit. Outcomes Each decision alternatives is associated with a set of possible outcomes for future states. for example when we decided to invest or not invest on something, each alternatives associated with possible outcomes.



There are five keys concept that becomes the foundation of Decision Analysis : Outcome Probabilities Each outcomes has outcome probability of occurence for each decision alternative that we take. The probabilities could or not be the same for various decision alternatives. Outcome Payoff Value For each decision alternative, has a numerical payoff value to us. If the decision we made and the outcome is not align with what we wanted (failed or not failed, etc) we must assign a numerical value to that circumstance. Expected Payoff The benefits or return that are assumed to be obtained when making a decision.



The Decision Under Risk Conditions Uncertain of future events, this event is used as a parameter to determine the decisions to be taken. The situation faced by decision-makers is to have more than one alternative course of action one alternative action Decision makers know the probabilities that will occur against various actions and their results by maximizing the expected return (ER) or expected monetary value (EMV).



Technical of Decision Under Risk Conditions Decision-making is a process that inseparable from solving a problem Imperfect information or not necessarily 100% correct makes decision making will face alternative solutions that have different levels of probability different. It is an analytic approach to selecting the best alternative or the best way to action. This technique is often used in managerial decision-making or production management.



Risk Decision Making Techniques



Expected Value



Expected Opportunity Loss (EOL)



Expected Value Of Perfect Information (EVPI)



1. EXPECTED VALUE The most frequently used criterion. The expected value for an action is the weighted average payoff, which is the sum of the payoffs for each action event multiplied by the probability of that event. The logical alternative is the one with the most significant expected value. Instead, what often happens is that the benefit is less significant than the expected value. This criterion is used because it can maximize the payoff in the long run (similar situations that occur repeatedly). repeated) can maximize payoff. If the situation is not repeated, then the use of the expected value is not appropriate.



2. EXPECTED OPPORTUNITY LOSS (EOL) To minimize the loss caused by choosing a particular decision alternative. The decisions recommended by the expected value and expected opportunity loss criteria are the same, and this is not a coincidence because these two methods always give the same results. Coincidence because these two methods always give the same results, so it is enough that one is used, depending on the objective. Only this criterion is highly dependent on accurate probability estimates.



3. EXPECTED VALUE OF PERFECT INFORMATION (EVPI) It is an extension of the EV and EOL criteria, or in other words, the information obtained by the decision maker can change the risk atmosphere into certainty (buying additional information to help the decision maker). EVPI is the same as the minimum (best) EOL because EOL measures the difference between the best EV of a decision in a risk and certainty environment.



Expected Payoff •Payoff is the difference between the market price and the exercise price, while profit is the difference between the payoff received and the premium paid (the purchase price of the option). The nominals could be positive or negative and might/not be measured on a monetary scale. •The value of the outcomes are ratio scaled for the individual making the decision. The term utility is often used in payoff value as not to suggest a monetary scale. However, utility has a variety of meanings, so we use payoff value in a general sense that can be measured on a monetary or non monetary scale.



aij: Alternative Value Pj : Probability Value



Key Point Looking for the alternatives with the highest pay-off



Example: The newspaper seller takes the newspaper in the morning and sells it, the selling price of the newspaper is Rp. 350 and the purchase price of Rp. 200 of the newspaper that does not sell well in the day does not have a price. From his notes the probability of newspapers selling daily: Prob1 = Sold 10 Exp = 0,10 Prob2 = Sold 50 Exp = 0,20 Prob3 = Sold 100 Exp = 0,30 Prob4 = Sold 150 Exp = 0,40



THE QUESTION: How many newspapers must be provided every day in order to get optimal profits?



Arrange a Pay Off Matrix Table



Calculate The Expected Pay Off



Conclusion: To get optimal results, the right alternative based on the highest expected result is 100 Exp Newspaper everyday.



SAPPK



EXPECTED VALUE METHOD



Expected Monetary Value (EMV) The Expected Monetary Value (EMV) method is a statistical concept analysis method that calculates the average future expenditure that may or may not occur. A positive EMV value indicates an opportunity, while a negative EMV value indicates a threat or threats that can harm the company. This EMV calculation is done by multiplying the probability/possibility of the risk occurring and the impact value of the risk if it occurs.



EMV = ∑Xi P(xi) X = Payoff for case-1 P(xi) = Probability of fulfilling the payoff (xi) ∑ = sigma/addition symbol



CASE STUDY



A housing developer sells a house at a price of 500 million/house. If the house is not sold (because the area has inadequate utilities/high crime rate) then Elaborate on what you is want to discuss. who wants to buy the house at there a contractor a price of 150 million/house. Meanwhile, the developer is thinking about how much inventory is ideal for the house he wants to sell. Based on experience from last year's period, where he sold the house for 500 days is as shown in the following table.



Payoff: payment/decision made State of nature: case



CASE STUDY A housing developer sells a house at a price of 500 million/house. If the house is not sold (because the area has inadequate utilities/high crime rate) then there is a contractor who wants to buy the house at a price of 150 million/house. Meanwhile, the developer is thinking about how much inventory is ideal for the house he wants to sell. Based on experience from last year's period, where he sold the house for 500 days is as shown in the following table.



Is known: - Selling price per house = 500 million - Selling price per house if not sold = 150 million -Prices of capital per house = 300 million -Profit per house = 500 million – 300 million = 200 million -Loss per house = 300 million – 150 million = 150 million



Nilai Probabilitas:



Tabel Payoff:



Tabel EMV: From the EMV table, it can be seen that the highest EMV is worth 3125. So



the



decision



optimal/ideal



taken



for



the



inventory



of



the



number of houses to be sold this year based on the EMV criteria is 8 houses for 200 days.



EXPECTED OPPORTUNITY LOSS (EOL) Main Point Choosing an alternative based on the minimum opportunity loss



Write Your Topic or Idea



The Implementation



A company is faced with a problem to choose three



investment alternatives, B and C. The profit obtained from the three types of investments depends on the market situation, namely lethargic (15%), normal (30%)



aij: Alternative Value Pj : Probability Value



and bright 55%).



THE QUESTION: Which type of investment to choose if the criteria are used opportunity loss ?



Choose The Highest Value Between All of The Probability



Calculate The Expected Opportunity Loss (EOL)



Conclusion: Investment Alternative C was chosen because it has the lowest EOL value, which is 1,500



EXPECTED VALUE OF PERFECT INFORMATION (EVPI) The Formula EVPI = EVWPI - EV



EVPI: Expected Value of Perfect Information EVWPI: Expected Value Without Perfect Information EV: Expected Value (EP/EOL)



The Implementation A company is faced with a problem to choose three investment alternatives, B and C. The profit obtained from the three types of investments depends



on



the



market



situation,



namely



lethargic (15%), normal (30%) and bright 55%). The marketing manager proposes to conduct market research at a cost of 2000.



Key Points: Looking for thresholds for costs to get the perfect information Comparing advantages When using perfect information and not



Add max rows by finding the largest value in each market condition



Calculate The EVPI EVPI = EVWPI - EV



80.000 35.000 165.000



Conclusion: An EVPI value of 1,500 means that the threshold for costs that can be incurred to obtain perfect information.



Decision Tree Method Any problem that can be presented in adecision table can be graphically representedin a decision tree. Most beneficial when a sequence of decisions mustbe made. All decision trees contain decision points/nodesand state-of-nature points/nodes. At decision nodes one of several alternatives maybe chosen. At state-of-nature nodes one state of nature will occur



FIVE STEPS OF DECISION TREE ANALYSIS 1. Define the problem 2. Structure or draw the decision tree 3. Assign probabilities to the states of nature 4.



Estimate



payoffs



for



each



possible



combination



of



alternatives and states ofnature 5. Solve the problem by computing expected monetary values (EMVs) for each state ofnature node



STRUCTURE OF DECISION TREE •Trees start from left to right •Trees represent decisions and outcomes in sequential order •Squares represent decision nodes •Circles represent states of nature nodes •Lines or branches connect the decisions nodes and the states of nature



D : Pengambil Keputusan a : Alternatif/ course of action 𝒙_𝟏 dan 𝒙_𝟐 : hasil/ payoff Oab : kejadian/ events (state of natures) Pab : probabilitas



CASE SUDY DECISION TREE 1 Today's Pertalite price is Rp 10,000. It was rumored in the news that tomorrow the price of Pertalite would rise to RP 14,000 or decrease to Rp 8,000 dengan probability 50:50.



EXPECTED MONETERY VELUES



Rp 11.000



EMV A = 8000 (0,5) + 14.000 (0,5) = Rp 11.000 EMV B = 10.000 (0,5) + 10.000 (0,5) = Rp 10.000



EXPECTED OPPERTUNITY LOSS



Rp 1.000



Rp 2.000



EOL A = 2000 (0,5) + 0 (0,5) = Rp 1000 EOL B = 0 (0,5) + 4.000 (0,5) = Rp 2000



CASE SUDY DECISION TREE 2 The government has alternative toll road construction projects of 200 km, 150 km or 120 km. If the development is successful, the government will get 80% profit, but if it fails, it will suffer a 20% loss.



EXPECTED MONETERY VELUES USD 370.000



USD 374.000



USD 356.000



EXPECTED OPPORTUNITY LOSS USD 26.000



USD 22.000



USD 40.000



EXERCISE MAKE DECISION TREE Mr. Natex wants to build a shoe factory, but he is confused about whether to build a large, medium or small factory, with the same probability value of favorable and unfavorable market. which one should Mr. NateX choose?



Thank you!