Sean O'Sullivan: Modelling - building a model of the problem that will generate the alternatives. Sep 29, 2009 1:00:53 PM EST
Sean O'Sullivan: Analysis - incorporating the functions and process that will evalueate each alternative. Sep 29, 2009 1:01:55 PM EST
Sean O'Sullivan: 'processes that will evaluate' Sep 29, 2009 1:02:16 PM EST
Sean O'Sullivan: And then we strike a BIG problem! Sep 29, 2009 1:02:58 PM EST
Sean O'Sullivan: which is -- many problems are very different from each other. for example ' what is the best roster for next month?' or 'how shoudl I spend my marketing budget for the year?' Sep 29, 2009 1:04:23 PM EST
Sean O'Sullivan: Each of these two problems require different modelling and analysis, and each will probably need a different DSS. Sep 29, 2009 1:05:23 PM EST
Showing posts with label Modeling and Analysis. Show all posts
Showing posts with label Modeling and Analysis. Show all posts
Wednesday, October 7, 2009
Wednesday, September 23, 2009
Simulation types page 167
Probalistic Simulation: One or more of the dependant variables are probalistic
Time Independant refers to a situation where it is not important to know exactly when the event occured
eg. Important to know that 3 units were sold in 1 day but not important to know when.
Time dependant, precise details of when the event happened are important.
eg. waiting lines in a que
Object Oriented Simulation: Using UML
Visual Simulation: Graphical display of computerised results and may include animation
- Discrete distributions: limited number of events or variables, finite number of values
- Continuous distributions: unlimited number of possible events
Time Independant refers to a situation where it is not important to know exactly when the event occured
eg. Important to know that 3 units were sold in 1 day but not important to know when.
Time dependant, precise details of when the event happened are important.
eg. waiting lines in a que
Object Oriented Simulation: Using UML
Visual Simulation: Graphical display of computerised results and may include animation
Simulation page 165
Simulation is the appearance of reality
A technique for conducting experiments (eg. what-if analysis) with a computer on a model of a management system.
One of the most commonly used methods DSS methods
A technique for conducting experiments (eg. what-if analysis) with a computer on a model of a management system.
One of the most commonly used methods DSS methods
Problem solving search methods page 162
Analaytical techniques: Use a mathematical formula to derive an optimal solution directly or to predict a certain result. Used primarily for solving structured problems, usually of a tactical or operational nature. areas such as resources allocation, inventory mgt
Algorithms: Step-by-step process for obtaining an optimal solution. Solutions generated and tested for possible improvements. The process continues until no further improvment is possible.
Blind Searching: In conducting a search, a description of a desired outcome may be given. This is called a goal. A set of possible steps leading from the initial consitions to the goal is called the search steps. Problem solving is odne by searching through possible solutions.
Heuristic Searching: Via knowledge, common sense, rule of thumb. Heuristics are the informal, judgemental kowl,edge of an application area that constitute the rules of good judgement in the feild. Domain KNowledge.
Algorithms: Step-by-step process for obtaining an optimal solution. Solutions generated and tested for possible improvements. The process continues until no further improvment is possible.
Blind Searching: In conducting a search, a description of a desired outcome may be given. This is called a goal. A set of possible steps leading from the initial consitions to the goal is called the search steps. Problem solving is odne by searching through possible solutions.
Heuristic Searching: Via knowledge, common sense, rule of thumb. Heuristics are the informal, judgemental kowl,edge of an application area that constitute the rules of good judgement in the feild. Domain KNowledge.
Multiple goals page 158
Managers want to attain simultaneous goals, some of which may conflict.
Difficulties of analysing multiple goals:
Difficulties of analysing multiple goals:
- Difficult to obtain explicit of the orgs goals
- Decision mkaer may change importance of assigned to specific goals over time or for different decision scenarios
- goals viewed differently at different levels of the org
- goals change in response to change in the org and environment
- relationships between alternatives and their role in determinning goals may be difficult to quantify
- complex probs are solved by groups of decision mkaers, each of whom has a personal agenda
- participants view the importance / priorities of goals differently
Linear Programming pge 154
Every LP problem is composed of:
- decision variables (who's values are unknown and are searched for),
- an objective function (a linear mathematical function that relates the decision variables to the goal, measures goal attainment, and is to be optimised),
- objective function coefficients (unit profit or cost coefficients indicating the contribution to the objective of one unit of a decision variable),
- constraints (expressed in the form of linear inequalities or equalities),
- capacities (upper, lower limits of variables), and
- input/output coefficients.
The components of DSS Mathematical problems page 151
All models are made up of four basic components. Mathematical relationships links these components together.
Result (outcome) variables
Decision variable (pg 58, 152)
uncontrollable variables (or parameters)
Result (outcome) variables
- Reflect the level of effectiveness of a system, ho well the system attains its goals. These variables are outputs.
- Considered dependant
Decision variable (pg 58, 152)
- Describes alternative courses of action. The decision mkaer controls the decision variables.
uncontrollable variables (or parameters)
- Factors that affect result varibales but are not under the control of the decision maker.
- Some of these variables limit the decion maker and therefore form what are called constraints of the problem.
- Reflect intermediate outcomes in mathematical models. eg. determining machine maintenance scheduling, spoilage, total profit, employee satisfaction
Decision Trees page 149
Alternative representation of the desicion table.
http://mindtools.com
Shows the relationships of the problems graphically and can handle complex situations in a compact form.
Can be cumbersome if there are many alternatives
http://mindtools.com
Shows the relationships of the problems graphically and can handle complex situations in a compact form.
Can be cumbersome if there are many alternatives
Treating Risk
The most common method for solving this analysis problem is to select the alternative with the greatest expected value. This approach dangerous becuase even an infintisimal chance or catastrophic loss can make the gains seem reasonable. But what happens if that small chance occurs?
Treating uncertainty page 149
Optimistic approach: assumes the best possible outcome of each alternative will occur and then selects the best.
Pessimistic approach: assumes that the worst possible outcome for each alternative will occur and selects the best of these.
Another approach is to simply assume that all states of nature are equally possible.
Pessimistic approach: assumes that the worst possible outcome for each alternative will occur and selects the best of these.
Another approach is to simply assume that all states of nature are equally possible.
Decision making under uncertainty
the decision maker considers situations in which several outcomes are possible for ech course of action.
does not know, cannot estimate the probability of occurance of the possible outcomes.
Insufficient information
involves assessment of the decision maker's attitude towards risk.
does not know, cannot estimate the probability of occurance of the possible outcomes.
Insufficient information
involves assessment of the decision maker's attitude towards risk.
Decision making under certainty
It is assumed that complete knowledge is available so that the decision maker knows the outcome of each course of action will be (deterministic environment).
The decision maker is viewed as being a perfect predictor of the future because it is assumed that there is only one outcome for each alternative.
Occurs most often with structured problems.
The decision maker is viewed as being a perfect predictor of the future because it is assumed that there is only one outcome for each alternative.
Occurs most often with structured problems.
Knowledge Clasification Page 143
Decision situations are often clasified on the basis of what the decision maker knows or believes about the forecasted results.
- Certainty
- Risk
- Uncertainty
A DSS can include multiple models
sometimes dozens, each of which represents a different part of the decision making problem.
Each model may either be native to the DSS or integrated, interfaced
Each model may either be native to the DSS or integrated, interfaced
Forecasting / Predictive Analysis
Essential for construction and manipulating models because when a deciion in implemeneted, the results usually occur in the future.
Ecommerce has created an immense need for forecasting and an abundance of available information for performing it.
Many orgs have accurately predicted demand for products and services using a variety of quantitive and qualitive methods.
CRM and revenue management systems rely heavily on forecasting techniques / predictive analysis
Ecommerce has created an immense need for forecasting and an abundance of available information for performing it.
Many orgs have accurately predicted demand for products and services using a variety of quantitive and qualitive methods.
CRM and revenue management systems rely heavily on forecasting techniques / predictive analysis
Subscribe to:
Posts (Atom)