This essay has been submitted by a student. This is not an example of the work written by our professional essay writers.
Every single project, in fact, will face unexpected conditions during execution phase. These conditions have different impacts, improving or reducing project performance. If those conditions, especially the "bad" ones - risks, are not well identified up front, they may potentially destroy the overall project execution. Project Risk Management is a world class process in identifying, assessing and developing plans to reduce or eliminate the risk impacts. In term of cost and schedule, this process prevents significant cost overrun and schedule slip by applying adequate contingency. One of Project Risk Management methodologies is Quantitative Risk Analysis. This study covers the combination of two AACE International Recommended Practices, Range Estimating and Expected Value, in conducting Quantitative Risk Analysis to determine the adequate contingency needed using simple Microsoft Excel® spreadsheet. Range Estimating method is used to identify and examine project uncertainties on each project schedule task durations and project cost components and Expected Value for project specific risks. Utilizing Monte Carlo simulation, the combination resulted on the S-Curve which was more skewed to the right. It reduces the project likelihood to achieve objectives with baseline cost and schedule is reduced and more contingency is needed to cover identified risks and uncertainties. The combination of both methods is very useful for comprehensive quantitative risk analysis. It will help project team to specify, identify, examine and encompass overall risks and uncertainties aspects which potentially impact to project outcomes and support management decision in determining adequate contingency requirement.
Successful project management requires strong leadership and organizational capability, good decision making process, achievable planning, effective communication, coordination and effective project control and monitoring of project execution. Effective combination of those requirements will guide the entire project team to carry out outstanding performance in achieving project objectives.
As Project Management definition above; cost, schedule, quality and safety are project key performance indicators or targets which determine successful project execution or not. During project execution phase, naturally, there are a lot of situations and conditions which were not predicted and potentially impact to project performances especially for cost and schedule. Those situations and conditions are defined as project risks and uncertainties.
In this case, project uncertainties are not only impact to worse outcome but potentially to better project outcome.
Realizing those potential risks and uncertainties, a project team requires a good project planning with comprehensive process to manage those risks and uncertainties. This process is defined as Project Risk Management.
In the following sections through the conclusion, the Author will discuss his perspective about:
General project risk management
How to prepare, develop and perform quantitative cost and schedule risk analysis
How to combine Range Estimating and Expected Value method in quantitative cost and schedule risk analysis
How to determine appropriate contingency level based on quantitative cost and schedule risk analysis result
Project Risk Management
Project risk management should have a good planning in order to make alignment among project team members and project stakeholders. Project risk management plan includes work flow and processes, risk factor screening matrix, risk owner determination including role and responsibly, monitoring and reporting. Usually each Corporation Company has their own standard project risk management process. This standard will be guidance, fit for purpose, for each project for developing specific risk management plan.
In this step, project team will identify all possible risks and uncertainties that may impact the project execution. A risk management workshop is usually held to review and capture potential project risks and uncertainties by brainstorming session. Depend on project size and complexity, beside project team members, other support teams, such as Supply Change Management, Government and Public Relation, Law, etc; experienced project teams which have similar project scope and experts are encouraged to participate in this workshop to gain values, alignment and perspectives.
After all possible project risks and uncertainties were identified, and then project team analyzes and assesses each risk qualitatively or quantitatively. For qualitative approach, project team will assign potential impact and likelihood of occurrence of each respected risk using risk factor screening matrix, shown on Figure 2, agreed in risk management plan. The combination between risk impact and likelihood will determine risk criticality category or level.
One of quantitative approach is conducting cost and schedule risk analysis by developing cost and schedule model, assessing and determining variability (uncertainties range) of each point estimate, define project specific risk events and then using Monte Carlo simulation to provide cumulative probabilistic output (more commonly known as S-Curve), shown on Figure 3. Monte Carlo simulation helps project team and stakeholders understand the range uncertainties and likelihood of achieving the planned outcomes. This simulation also can provide sensitivity (tornado) diagram which represent project risk drivers (priority) of respected point estimate.
Since not all risks have high (critical) level impact to project outcome, project team need to prioritize all analyzed risks. Then project team need to develop risk action plan including specific mitigation plan, associated timing and resources and assign risk owner to all prioritized high level impact risks. Then this process continues to action plan implementation.
From Process Map for Risk Management, showed on Figure 1, we can see that project risk management is not only event based process but it is a continuous process that will improve project understanding on risks and uncertainties that potentially impact to project outcome. Project risk management is most effective if it is monitored, controlled and adjusted if required.
Quantitative Cost and Schedule Risk Analysis
In project world, risks and uncertainties are translated as contingency. Contingency refers primarily to the amount of quantity of funds, time or other financial resources that is required to be allocated at and above the previously designated cost and schedule estimate amount to reduce the risk of overruns to an acceptable level for the financially.
As mentioned above, quantitative cost and schedule risk analysis utilizes Monte Carlo simulation. Monte Carlo is one of technique and tool for risks and uncertainties analysis. This tool is used for generating probabilities through random sampling or iteration of all possible value of uncertainties and risks. The simulation result accuracy depends on number of iterations. The accuracy can be illustrated as follows:
This method can support project team in quantifying project risks and uncertainties and determining appropriate contingency level. Using mathematical techniques and models, quantitative risk analysis numerically comes up with cumulative probabilistic result. This simulation result communicates potential risks and allows the management to select appropriate risk (contingency) level based on company approach. Without this valuable information, project's invested cost and schedule could be in corrected and causing underestimate which possibly came up with project overruns. This condition can possibly cause the project to require funding amendment to achieve project outcomes.
There are several steps in conducting quantitative cost and schedule risk analysis:
Developing cost and schedule models.
Using Work Breakdown Structure (WBS) and Cost Breakdown Structure (CBS) as starting point are best practices in analysis model development. The cost and schedule models should represent all project scopes.
Determine associated risks and uncertainties variables of each cost or schedule elements
From developed models, we determine associated risks and uncertainties on each cost and schedules elements. AACE International has published two Recommended Practices (RP) in conducting risk analysis and contingency determination. They are using range estimating RP 41R-08 and expected value RP 44R-08.
Assign Probability Distribution Function for each cost or schedule elements
Since we will use Monte Carlo simulation, we need to define possible values from risks and uncertainties variables determined from previous step. In this case, we have to define probability distribution of each cost and schedule elements. Triangular distribution is the most common probability distribution type used for cost and schedule variables.
Assign correlation factor between related or depended cost or schedule elements
Since there are possibilities of dependency of some cost or schedule elements, project team need to assign correlation factor between two or more elements. Correlation factor is defined from -1.0 to +1.0, where 0 indicates no correlation. Without appropriate data, correlation factor is quite difficult to determine.
Analyze the models using Monte Carlo simulation
Monte Carlo simulation will generate cumulative probabilistic curve (S-Curve) which shows alternative cost or schedule probabilities based on risks and uncertainties variables inputted. Besides producing this curve, Monte Carlo also able to communicate cost and schedule elements which drive uncertain result. This information is provided by Tornado Diagram, shown on Figure 4.
Analyze simulation result and generate report
The S-Curve communicates how likely our current (base) cost and schedule estimates are to over-run or under-run and how much contingency is needed to justify level of confidence of project to finish on time and on budget. Each company has own target and willingness to accept risks. As Tornado Diagram provides prioritized risk drivers information, project team need to develop mitigation plan to reduce amount of risks and uncertainties impact and likelihood of occurrence of respected risk drivers.
As mentioned previously, quantitative cost and schedule risk analysis should be conducted periodically. In line with mitigation plan progress, we expect that several risks and uncertainties impacts already have been reduced. By having continuous risk analysis and mitigation plan, it will maximize a project's chance to be delivered on time, on budget and safely.
Cost and Schedule Risk Analysis for Appropriate Contingency Determination using Range Estimating and Expected Value Methods
As described above, after we developed cost and schedule models, the next step is determining associated uncertainties and risks that as per experiences and data may influence and impact to project execution. AACE International recommends two methodologies in determining associated risks and uncertainties; they are Range Estimating and Expected Value.
In Range Estimating method, the project team and other workshop participants will determine the ranges of each cost and schedule elements based on their knowledge, experiences of similar projects and any available data and/or benchmarking information. Utilizing triangular probability distribution function, the determined range will be iterated in Monte Carlo simulation to generate S-Curve. The Expected Value method is used to describe specific project risks that may impact project in specific condition or period of time.
The Expected Value in its most basic form can be expressed as follows:
If triangular probability distribution function is used for Range Estimating, Expected Value uses Bernoulli probability distribution function for representing the probability of risk occurring of project specific risk.
In this section, the Author will illustrate the combination of both AACE International Recommended Practices by using simplified real project cost and schedule estimate data for developing the model, ranges and risk drivers with modified values and also Microsoft Excel with Crystal Ball as Monte Carlo simulation software.
4.1. Quantitative Schedule Risk Analysis with Range Estimating and Expected Value Method
The first step in performing quantitative schedule risk analysis is developing schedule model. The best approach in developing this model is by utilizing the Critical Path Method (CPM) schedule. By selecting critical path and near critical path tasks, we will be able to develop the model. The important thing in developing the schedule model, it should represent the entire project scopes. Developing a schedule model network diagram can help all workshop participants in understanding the model and guide the discussion in determining the uncertainties range and project specific risk of each schedule task.
Since we use Microsoft Excel spreadsheet in this risk analysis, manual calculations and formulas are used to describe relationship between tasks and determine start and finish date of each task. To prevent logic change during risk analysis discussion, the schedule model should be understood and agreed ahead by all parties who involve in risk analysis.
Once schedule model is developed, the next step is determining the uncertainties range of each schedule task. In AACEI RP 41R-08, this method is called as Range Estimating. The risk workshop participants will discuss and determine appropriate uncertainties range, shown in Table 2 with yellow highlights, of each task based on their experiences, judgment and available data. Remember to capture all related discussion during determining the range as workshop evidences and valuable information in the future. As described above, the next step, we have to determine Probability Distribution Function (PDF) for all task ranges and specific risks. In this project, we use triangular PDF for all task ranges, as shown in Figure 6.
Once all schedule task ranges are done, then we need to check and determine additional project specific risks which potentially impact to specific project schedule task. In this project example, there are two project specific risks identified and all of them relate with procurement process. In Indonesia, all Oil and Gas Companies have to comply with Indonesian Executive Agency for Upstream Oil and Gas Industry (BPMIGAS) regulation in procurement process (PTK-007). Both project specific risks are retender on material (equipment) and contract procurement processes. Based on data from Supply Change Management (SCM) team, retender on material procurement will add 30 days with likelihood of occurrence is 15% and retender on EPCI contract procurement will add 60 days with likelihood of occurrence is 20%.
We use "Yes-No" PDF for both specific risks represent likelihood of occurrence, as shown in Figure 7. Then material procurement retender risk is applied and tied-in to "Major Equipment & Bulk Procurement Process - ID 4" and also EPCI contract procurement retender risk to "EPCI Contract Procurement Process - ID 6", as shown in Table 3 - the same PDF is applied for Retender Material.
After assigned correlation factors, we run the Monte Carlo simulation and come up with S-Curve as shown in Figure 8. From this figure, we can see the likelihood of achieving original project completion date (baseline schedule) is only around 25% of confidence level. If we compare this simulation result with simulation result without project specific risks, as shown in Figure 9, we can see there is approximately 14 days difference on P50 confidence level. It means that project specific risks impact to overall project duration, even though in this project example, the contribution is not significant. It is only 2% of total baseline duration.
Using the S-Curve, the decision makers will be able to determine the contingency needed to achieve an expected confidence level. Each company typically has an expectation setting on level of acceptable risk tolerance or confidence level. The contingency level is determined by the difference between the acceptable confidence level and project baseline. If the Mean, represent as Expected Value, is selected as the acceptable level then the project schedule contingency is 72 days.
Another advantage in Monte Carlo simulation is identification of high priority risks and uncertainties which drive uncertain simulation result. This information can be illustrated by Tornado Diagram, as shown in Figure 10.
In this project example, uncertainties range of schedule tasks drive the project finish date. Since project specific risks have no significant impact to finish date, both risks are not considered as high priority project risk drivers. From Tornado Chart above, project team need to develop mitigation plan to reduce amount of uncertainties impact of respected risk drivers.
4.2. Quantitative Cost Risk Analysis with Range Estimating and Expected Value Method
Similar like schedule risk analysis, the cost model can be developed from high level Cost Breakdown Structure. Each cost element is formed by the combination of unit amount (scope) and unit rate. The next step, we will determine uncertainties range of each cost element scope and rate and the result is as shown in Table 4. The triangular PDF is also applied to all cost elements.
As we know that schedule slip or delay will contribute in increasing the cost. Several cost elements may depend on specific schedule tasks duration. It means that each cost element risks depends not only its components (scope and rate) but also from on respected duration uncertainties. The cost components uncertainties already determined by range estimating method above. Since specific schedule duration uncertainties potentially impact to specific cost elements, they can be utilized as project specific risks for cost risk analysis. From this project example, there are two schedule task durations are identified and impact to several cost elements as shown in Table 5.
Both project specific risks information can be gathered from Schedule Risk Analysis simulation result. The "Total Site Installation Duration" is defined as all activity durations from "Field Fabrication and Assembly" until "Commissioning & Start Up". As Monte Carlo simulation, this project specific risk comes up with S-Curve as shown in Figure 11.
From S-Curve above, we can calculate the difference between baseline duration and major percentiles, as shown in Table 6 - yellow highlighted and use Triangular PDF for simulation later, as shown in Figure 12. The same method is applied for 2nd specific risk by using Figure 8, the total project duration.
After assigned correlation factors, we run the Monte Carlo simulation and come up with S-Curve as shown in Figure 13. From this figure, we can see the likelihood of achieving original project completion date (baseline schedule) is only around 15% of confidence level. If we compare this simulation result with simulation result without project specific risks, as shown in Figure 14, we can see there is approximately US$ 1.2 million difference on P50 confidence level. It means that schedule duration uncertainties (specific risks) allocate significant impact to total project cost, with additional contingency 8%.
From the Tornado Diagram, as shown in Figure 15, we can see that schedule duration uncertainties become the main risk driver of total project cost uncertainties, beside several cost components. This condition is concurred with above statement that "schedule slip will contribute in increasing the cost".
The above simulation results show the implication of the combination between Range Estimating and Expected Value methods in conducting quantitative cost and schedule risk analysis. The combination of both methods will help project team to specify, identify, examine and encompass overall risks and uncertainties aspects which potentially impact to project outcomes and support management decision in determining adequate contingency requirement.
The quality of quantitative risk analysis result depends on workshop participants' knowledge, experiences of similar projects, judgments and any available data and/or benchmarking information. Poor quality analysis leads to inaccurate decision making.
Realizing this condition, the quantitative risk analysis should be conducted periodically to ensure the validity of all risks information and effectiveness of mitigation plan. Effective Risk Management plan will improve the likelihood the project in achieving its goals.