@RISK for Project. p. 1 of 8

Transcription

1 Product Section: Risk Analysis Operating System: Windows - Add-in for Microsoft Project Palisade Product Number: 1007 Standard, 1007P Professional Retail Price with Maintenance: $950 Standard, $1,550 Professional p. 1 of 8 Add Risk Analysis to Project is the Risk Analysis and Simulation add-in for Microsoft Project. Add the power of Monte Carlo simulation to your Project for Project features all the power and ease of use for Microsoft Excel. Simply replace uncertain values in your project functions which represent a range of possible values. Select bottom-line fields like Finish Date, Project Duration, or Total Cost as outputs, and start a recalculates your project hundreds or even thousands of times, each time selecting random numbers from functions you entered. The result: distributions of possible outcomes and the probabilities of getting those results. This not only tells you what could happen in a given situation, but how likely it is that it will happen. Sensitivity and Scenario analyses will help you identify critical situations that could seriously impact your project. Account for unforeseen risks in your projects for for Project comes in two editions Standard and Professional to allow you to select the feature set which best meets your needs. Risk Analysis Risk Analysis, in a broad sense, is any method qualitative and/or quantitative for assessing the impacts of risk on decisions. Myriad Risk Analysis methods are used that blend both qualitative and quantitative techniques. The goal of any of these methods is to help the decision-maker choose a course of action, given a better understanding of the possible outcomes that could occur. You might wonder if what you do would be suitable for Risk Analysis. If you use data to solve problems, make forecasts, develop strategies, or make decisions, then you definitely should consider performing Risk Analysis. Risk Analysis Risk Analysis is a quantitative method that seeks to determine the outcomes of a decision as a probability distribution. In general, Risk Analysis encompasses four steps: 1. Develop a Model First, define your problem or situation in Microsoft Project. 2. Identify Uncertainty Next, determine which inputs in your model are uncertain, and represent those using ranges of values probability distribution functions. Identify which result or output of your model you want to analyze. 3. Analyze the Model with Simulation Run your simulation to determine the range and probabilities of all possible outcomes for the outputs you ve identified. 4. Make a Decision Armed with complete information from your analysis, and your personal preferences, make your for Project helps with the first three steps by providing a powerful and flexible tool that works with Project to facilitate model building and Risk Analysis. The results for Project generates can then be used by the project manager to help choose a course of action.

2 p. 2 of 8 Step 1: Develop a Model The first step is constructing a model in Project that represents your situation. Because you are working in Project, you are virtually unlimited in what type of Project you can model. You might be constructing a new skyscraper and want to determine what the budget and deadline should be. Or, you may be phasing in a new computer system within your organization and need to assess the impact on productivity. Whatever your for Project can help! Step 2: Define the Uncertainty Using Model Definition and Define Distribution Windows Next you need to represent uncertainty in your Project model. You probably base most decisions on whatever data you have on hand historical costs, competitors prices, vendor estimates, etc. But how often do you have full, complete information? Prices change, weather intervenes, people are out of work, costs rise. By using probability distribution functions to represent a range of possible lets you take that uncertainty into account. Simply select a cell whose value you are unsure of and replace its value with one of the probability distributions below: Beta Beta General Beta-Subjective Binomial Chi-Square Cumulative Discrete Discrete Uniform Error Function Erlang Exponential Extreme Value/Gumbel Gamma General Geometric Histogram Hypergeometric Int Uniform Inverse Gaussian/Wald Logistic Log-Logistic Lognormal Lognormal2 Negative Binomial Normal Pareto Pareto 2 Pearson V Pearson VI PERT Poisson Rayleigh Student's t Triangular Triangular General Uniform Vary Weibull For example, if you have a resource field cost estimate with a most likely value of $10,000, but you know it can t go below $7,000 or higher than $15,000, you can represent that uncertainty with a Triangular distribution using the function RiskTriang(7000, 10000, 4.1 for Project makes entering your probability functions easy. Just click the Define Distribution icon to use Define Distribution Window for entering your probability distributions graphically (see below for details). Or you can simply type them Functions column of Project once you get accustomed to the syntax. Once you ve entered your distribution functions, select the bottom-line field or fields whose values you are interested in (such as Total Cost, Project Finish Date, or Project Duration) and click the Add Output button in toolbar. This adds a RiskOutput function to the chosen field, allowing you to name, move, and manage your outputs easily. Open Model Definition Window to see a clear list of your inputs and outputs at a glance. You can also perform distribution fitting, correlate inputs, and open a Define Distribution Window from Model Definition Window. Model Definition Window lets you see elements at once.

3 Visualize Your Uncertainty with allows complete graphical selection of your probability distribution functions using RISKview, Palisade s distribution previewing companion. Now entering your distributions is easier than ever because you can see them, edit them, and apply them with a few clicks of the mouse. RISKview comes fully integrated with every copy 4.1 for Project, and appears as a convenient pop-up Define Distribution Window right over your project model. You can view graphs, parameters, and statistics of distribution. All graphs include adjustable delimiters and probability bars for viewing calculated probabilities and other statistics directly on graphs. Simply slide the delimiter on a graph to see calculated probabilities both on the graph and in the linked statistical report. RISKview even lets you overlay graphs of different distributions for comparison, change parameters, and draw your own curves! When you draw your own curves using RISKview s Distribution Artist, RISKview can find the standard probability distribution which best fits that curve. RISKview can also directly link to any distribution in your spreadsheet model for instant updating. Defining your uncertainty couldn t be more intuitive! RISKview pops up right over your Project model! p. 3 of 8 To further speed up the process of defining uncertainty, RISKview features a Distribution Palette of thumbnail sketches of all available distribution types. Just click the thumbnail of the distribution you want. And, you can enter dates as parameters in your distribution functions. For instance, you could define a Normal distribution with a mean of 10/1/2002 and a standard deviation of 4 days. Fit Distributions to Data with BestFit There may be times when you have historical data on a particular input in your model. For example, you may have price data on the costs vendors charge for outsourced jobs. You would like to use this data to represent uncertainty in your model, but how? With BestFit, Palisade s distribution fitting tool, it s easy! BestFit is fully integrated with the Professional version 4.1 for Project. Just read your data into Fitting Window, and click the Fit button. BestFit will find the distribution which best describes your data. You can then apply your fitted distribution to your Project model. BestFit gives you multiple graphing options and full statistics for your fits. BestFit ranks dozens of distribution functions against your data, and provides comprehensive results. Four types of graphs are available to assess accuracy of the fit, and there are complete statistical reports and goodness-of-fit data. Choose from Comparison, Difference, Probability-Probability, or Quantile-Quantile graphs. Like RISKview, BestFit includes sliding delimiters and probability bars on all graphs. Graph types can be easily formatted be using toolbar icons or rightclick menus, and may be exported to Excel in native Excel format for further modification. But it doesn t stop there. BestFit has a host of advanced features for the power user. BestFit uses three advanced fitting algorithms to optimize its fits Chi Square, Anderson-Darling, and Komolgorov-Smirnov. You can read in data sets with up to 100,000 points, in sample, density, or cumulative format. You can fit multiple data sets in a single project, and specify which predefined distributions to fit to. BestFit allows full control over Chi-squared calculations, including equal interval binning, equal probability binning, and full custom binning. It will also perform the RMS (root mean square error) test for cumulative and density data.

4 p. 4 of 8 Step 3: Simulate and View All Possible Outcomes! Once your model is set up, click the Simulation Settings button if you want to customize the parameters of your simulation. You can specify the number of iterations recalculates the project model), update the project in real-time numerically or graphically is simulating, control the convergence criteria, and choose Monte Carlo or Latin Hypercube sampling. This gives you complete control over your simulations. Or, use the default settings will automate everything for you. Then click the Simulate button and recalculates your project hundreds or thousands of times! Each selects random numbers from functions you entered and records the resulting output. Each recalculation shows a possible combination of uncertain values or a scenario that could occur. At the end of the simulation, you have a whole range of possible outcomes, and the probabilities of them occurring! Your project has gone from representing one possible scenario to representing all possible scenarios! Try doing that with Microsoft Project alone. Analyze Your Results in Results Window When your simulation is complete, all the results appear instantly in Results Window. The Results Window features an Explorer-style list of inputs and tabbed reports. Extensive graphing options, comprehensive statistics, full data, and advanced analyses like Sensitivity and Scenario analyses are available with the click of a button. Dazzling Graphs One of the strengths of Monte Carlo simulation is that it produces enough data to create accurate graphs. Histograms, cumulative curves, area and line graphs, and more are available 4.1 for Project. Graphs are created with simple toolbar clicks or right-click menus. Select the type of graph, then customize it by selecting colors, changing scaling, creating new titles and axis labels, and more! Do you have a range of outputs you d like to graph? Simply highlight the range in the Explorer list and select Summary Graph to see your risk over time. You can even overlay multiple variables on the same graph for comparison. All graphs may also be displayed in Excel format for further enhancement. These Excel format graphs recreate graph in native Excel chart format, giving you access to all of Excel s charting capabilities. You can also export all graphs directly to your Project s various graphing options allow you to quickly and easily present possible outcomes for any situation to others! Results Window gives you countless graphing and reporting options! All graphs include summary statistics in the same window for easy reference. They also include probability bars and sliding delimiters, so you can answer questions like What are the chances of finishing on time? simply by sliding a bar to the deadline date mark on your cumulative curve and reading the probability result can also create a summary report of results using the Quick Report command. With a single will generate a one-page, preformatted report in Excel containing a histogram, cumulative curve, Tornado graph, and summary statistics all set up and ready to print! This saves valuable time by doing the formatting for you. Want to see how your graph changes as the simulation will display and update any graph in real-time while your simulation is running. You can start, stop, and control the frequency of updates.

5 p. 5 of 8 Risk Results in your Gantt 4.1 for Project displays your risk simulation results fully integrated with - and in the same format as - Project's native Gantt charts. New bars indicate the range of possible values for uncertain variables, and display sensitivity and critical index information. See how your risks impact your entire project - at a glance! The command "Create Risk Gantt Chart" is used to place Risk Gantt Charts and Tables in Microsoft Project. The Risk Gantt Chart shows, by task, earliest, expected and latest start and finish dates, along with critical index and sensitivity analysis results. The same information is available in tabular form right beside the Gantt chart. Identify Critical Factors with Sensitivity and Scenario performs two additional analyses that identify your risk factors: Sensitivity Analysis and Scenario Analysis, both accessible by clicking their tabs in the Results Window. Sensitivity Analysis determines which input Integrated risk results in the Project Gantt chart make interpreting your simulations a snap. distributions have the biggest impact on the outputs. The results can be displayed as an easy-to-interpret Tornado chart, with longer bars at the top representing the most significant input variables. You can choose from Rank-Order Correlation or Multivariate Stepwise Regression to perform Sensitivity Analysis. Sensitivity Analysis allows you to zero in on the factors in your model which contribute the most to your risk! Scenario Analysis identifies combinations of inputs or scenarios which lead to output target values. This lets you make observations such as, When this project tends to go overbudget, significant inputs are high material prices, delays in finishing Phase 2, etc. Accurate Statistics and Data Click the Detailed Statistics button to see statistics for all outputs displayed in a spreadsheet-like format for easy viewing. All the complete statistics you would expect are here. You can also enter target values and find the likelihood of achieving them, similar to using the sliding delimiters on the graphs. Another click and you can see all data for inputs and outputs for every iteration of the simulation. You can cut and paste data and statistics to other applications, or generate a full statistical or data report in your Excel for further analysis. Plus, you can export statistics directly to Project after a allows full access to the information of your simulation! Customizable Reports The best analysis always has to be presented to others. To help with provides extensive capabilities for reporting on both your model and simulation results. To start, graph or report can be pasted into Excel, Project, or other applications. In comes with a report generator that will quickly create a report in Excel with graphs and statistics on your simulation. Step 4: Make a Decision s comprehensive results and your own personal thresholds for risk, you are now ready to make a well-informed decision. By accounting for all possible gives you the tools to make the best possible decision in any situation!

6 p. 6 of 8 Advanced Features a True comes with a number of advanced features that make it the only choice for top decision-makers world wide. Critical and Cruciality Indices Critical indices determine when a task falls along a critical path during a simulation. Critical tasks are tasks that fall on the critical path a sequence of tasks that determines the total time for the project. But when you add uncertainty to a project, tasks may or may not be critical depending on the values sampled during the for Project automatically records the critical indices for all tasks as a percentage of time the task is critical during a simulation, allowing you to better identify the critical tasks in your project. for Project cruciality index factors in both critical indices and sensitivity analysis for the most accurate measure yet of a task's importance. Multiple features the RiskSimTable function, which allows multiple simulations to be run back-to-back. This lets you change only select inputs in your model from simulation to simulation, so you can compare the effect of a variable or variables on your outcome. For example, you may with to see what impact charging different contractors has on your project cost. You can even graph the results from multiple simulations on one chart! Correlate Your Inputs for More Accurate Models In real life, inputs are seldom independent. When interest rates are high, mortgages are low, for example. If you don t take these relationships into account when building your model, you risk dangerously inaccurate results. Model Definition Window provides an easy way to correlate inputs in your model using a simple matrix format. Select the inputs to correlate, right click, and choose Correlate. Enter in correlation coefficients, click Apply, and the correlations are automatically written to your Project model. Correlate your inputs in Model Definition Window. Enhanced VBA Support for Custom lets you write your own custom applications in Project that can perform virtually function. Harness the power s Monte Carlo simulation engine for your company s specific needs. s extensive reporting options, Sensitivity analysis, Scenario analysis, distribution fitting, and more, all in your own custom Project program! Example files are included demonstrating how to use these commands. Plus, you can run macros before, during, or after a simulation run. Risk Categories Many times the uncertainty for a field s value will be common to similar tasks. Consider a software development project. You may wish to assign a common variation to the duration of all programming tasks, and a different, but also common, variation to the duration of all testing tasks. The Risk Categories feature lets you easily define a + or percent change to your estimated values in just a few 4.1 is available in two editions to meet your risk analysis needs: Standard and Professional. In addition to BestFit, described Professional adds a host of analytical power features to your arsenal. Probabilistic Branching Probabilistic branching allows a project to branch from one task to any number of other tasks during the simulation. Each of the task groups that could be branched to has a probability value. For example, upon completion of a design task, there may be a 10% chance that the bad weather construction tasks will be followed and a 90% chance that the standard construction tasks will be followed. Account for chance events when planning your projects!

7 If/Then Conditional Modeling lets you build logic into your project. p. 7 of 8 If/Then Conditional Modeling If/then conditional modeling allows you to check if certain conditions are true during a simulation, and if they are to change values in the project or perform branching. For example, the user could enter a condition such as If Construction[Finish] > 1/1/05 Then Cleanup[Fixed Costs]=20000 or If Construction[Duration] > 50d Then Branch=Temporary Office Space Procurement. If a condition was true in an iteration of the will make the appropriate change to the project so the results include the conditional effect. You can even distribution functions directly in the IF/THEN dialog as a result of certain conditions. Save time by not having to define a global variable when you have just a few conditions to model. Control values and branching between tasks using logic statements that you define! Probabilistic Calendars Often you will need to model the probability of work stoppages due to weather, strikes, or other unforeseen events. By setting up Probabilistic Calendars you can apply working and non-working probabilities to any calendar event in Project! Global Variables Global variables can be used to hold values that can be referenced calculations in your project. Variables may be defined distributions, single fixed values, or even If/Then conditions that reference other variables. Variables are not attached to a specific task or field; they are attached to the project as a whole. A variable Bad Weather, for example, could be defined to occur 20% of the time. Changes in your project estimates could then be triggered when Bad Weather occurs. Field and Macro Pro for Project also supports field and macro references in distributions. Task[Field] references and VBA macros can be used as arguments to distribution functions. With this new distribution functions can be entered with non-numeric arguments, such as RiskNormal(Design[Duration],5). This allows values for distribution arguments to change each iteration of a simulation, as values in referenced fields change. Parameter Entry Tables There may be times when you wish to assign a distribution to a particular field throughout the project, and you wish to allow other users to change the parameters of that distribution. Parameter Entry tables make this easy! For example, in a few clicks you could assign a Normal distribution to all duration fields your entire model, and allow others to make changes to the default parameters based on their expertise and data. Become for Project for Project comes with a wide variety of resources to help you get started and overcome problems. A comprehensive audio/visual tutorial walks you through the features and demonstrates how to set up and run a model, step-by-step. manual is written in straightforward, plain English and even provides background information on Risk Analysis. The entire manual is also available electronically s Help features a comprehensive, context-sensitive Help file to zero in your question. You can search on a particular topic, or simply ask a plain English question using the How Do I... feature! Finally, example files are provided that can be used as templates for building your own model. Palisade also offers live training for Project. You can attend a regional seminar, or a Palisade trainer can fly to your office for customized, on-site training. Visit or call our Training Department at RISK.

8 @RISK for Project Applications Construction Engineering Computer Systems Installation Environmental Manufacturing Aerospace Pharmaceutical Mergers & Acquisitions Mortgage Pricing Operation Research Analysis Petroleum and Mining Resource Evaluations And Much More! p. 8 of 4.1 for Project Features True Project Add-In Toolbars for Most Commands Office-style Model Definition Window 38 Distribution Functions Fully Integrated RISKview Fully Integrated BestFit (Professional version only) Distribution Palette Support for Date Formats Correlation of Inputs Customizable Simulation Settings Convergence Monitoring Full Statistics Report Full Data Report Reporting in Excel Reporting in Project Target Values Histogram, Area, Line, and Cumulative Graphs Fully Customizable Graphs Summary Graphs Multiple Summary and Overlay Graphs Sliding Delimiters on all Graphs Real-time Updating of Graphs Risk results in Gantt Charts Graph in Excel One-step Quick Reports Multiple Simulations Sensitivity Analysis - Multivariate Stepwise Regression and Rank Order Correlation Tornado Graphs Scenario Analysis Statistics Functions Macros - Execute During Simulation Macros Customize with VBA Probabilistic Branching (Pro only) If/Then Conditional Modeling (Pro only) Global Variables (Pro only) Field and Macro References (Pro only) Parameter Entry Tables (Pro only) Comprehensive Help File and Tutorial System Requirements: IBM PC Pentium or higher; Windows 98 or higher; Project 2000 or higher; 16MB RAM or more Pricing and Maintenance: $950 Standard, $1,550 Professional All prices include one year of maintenance. Maintenance includes free, unlimited technical support and free version upgrades. Network, corporate, and academic pricing available. Other Versions for Developer s Demo: Free demo CD available Recommended Books: Risk and Decision Analysis in Projects; Practical Risk Assessment for Project Management