**5. Results and discussion**

The deterministic project approach traditionally requires that the values for all input data be known exactly. But data in many real manufacturing projects cannot be precisely given. The stochastic approach is based on the replacement deterministic data with random variables. Important studies to stochastic variables incorporated in the data envelopment analysis can be found in (Sengupta, 1982, 1987, 1990, 1997, 1998, 2000, Cooper et al, 1998; Huang & Li, 1996; Morita & Seiford, 1999; Sueyoshi, 2000, as cited in Azadi & Saen, 2011).

Risk Analysis of the Waste to Energy Pyrolysis Facility Designs for City of Konin, in Poland, Using SimLab®

considered at least 10,000 times.

(Bieda, 2007).

using of the SRC is shown to be extremely satisfactory (Bieda, 2010).

*Regression Coefficients*, SRC) and Figure 17 (sensitivity analysis - *Cobwebs plot* based on the *Standardised Regression Coefficients*, SRC) (Bieda, 2011). Based on the economic feasibility model presented in (Liberman, 2003), in this study used uniform distributions. Figure 15 and Figure 16 shown the results from the sensitivity analysis. The performance of the SRC is shown to be extremely satisfactory when the model output varies linearly or at least monotonically with each independent variable. MC analysis-simulation is the only acceptable approach for U.S. Environmental Protection Agency (EPA) risk assessments (Smith, 2006). Because all of the parameters of the economic model are independent, the

SimLab® is didactical software designed for global uncertainty and sensitivity analysis, developed by the Joint Research Centre of the European Commission and downloadable for free at: http://simlab.jrc.ec.europa.eu (Simlab*,* 2004). The sampling techniques available in SimLab® are FAST, Extended FAST, Fixed sampling, Latin Hypercube, replicated Latin Hypercube, Morris, Quasi-random LpTau, Random and Sobol (Saltelli et al, 2004). SimLab® can also run models built in Microsoft Excel®. Using the SimLab® in order to determine the most relevant parameters sampling presented by sensitivity analysis, after selected the Monte Carlo (MC) simulation, sampling method, the optimal number of executions is

There are available various commercial software packages in order to conduct the risk analysis using MC simulation. Among them, Risk® and Crystal Ball®, developed by Palisade Corporation and Decisioneering, respectively. Risk® was originally designed for business application and is easy to use without a need for extensive statistical knowledge and modeling capacity (Sonnemann et al, 2004). Crystal Ball® is a simulation program that helps analyze the uncertainties associated with Microsoft Excel® spreadsheet models by MC simulation (Sonnemann et al, 2004). Another function of the Crystal Ball® is the sensitivity analysis

According to Hullet (2004) the estimate of project cost risk can be made more accurate and better understood if the sources of risk are disaggregated into those that affect time and those that affect the burn tate per unit time. The schedule risk and cost risk analysis have been conducted in Microsoft Excel® and Crystal Ball®. In conclusion, Hullet (2004) muses that cost risk analysis that explicitly incorporates schedule risk analysis results, merging them with burn rate risk information in the estimates of cost risk that are more accurate than the typical approach. In the opinion of Leach (2005) anyone who is serious about realistically forecasting project schedules, in other words, truly managing projects, rather than just monitoring them, should be using MC simulation software to plan and analyze projects stochastically. Stochastic simulation (often called MC simulation) allows to capture and understand the uncertainty inherent in the project. Anderson (2005) in paper presented to the Denver Crystal Ball Conference in 2005 discusses the results of the using the MC simulation instead the analytic approach in the nuclear power plant steam generator repair/replacement cost/benefit analysis (before nuclear power plant steam generator replacement decisions have never included a MC simulation) and the strengths of the weaknesses of using Crystal Ball® and MC simulation. In conclusion, Anderson (2005) seems to think that MC approach was clearly appropriate to fully assess the impact of any

Toolpack 251

Projects involve risk by nature (Zwikael & Ahn, 2011). Reducing the level of risk is extremely important in projects, and indeed results of this study suggest that project managers often use risk management planning practices, consistent with previous studies (Kerzner, 2009; Ahmed & Kayis, 2007; Voetsch, 2004; Zwikael, 2004, as cited in Zwikael & Ahn. 2011). The Project Management Body of Knowledge (PMBOK) defines risk management as one of nine project knowledge areas, alongside other topics such as scope, schedule, quality, and cost management (PMI Standards Committee [PMI], 2008). In some project contexts, risk management is perceived as a separate activity (Zwikael & Ahn, 2011).


When applying for European Union (EU) funding of projects, cost-benefit analysis (CBA) practitioners need to prepare comprehensive investment appraisals following the latest guidelines on CBA provided by the European Commission (2008). Since this Guide includes the need to conduct a proper risk analysis, partly through sensitivity analysis (Evans & Kula, 2011).

The purpose of sensitivity analysis is to determine the relationships between the uncertainty in the independent variables used in an analysis and the uncertainty in the resultant dependent variables. Sensitivity refers to the amount of uncertainty in a forecast that is caused by the uncertainty of an assumption as well as by the model itself. Sensitivity analysis can be used to find "*switch points*" -- critical parameter values at which estimated net benefits change sign or the low cost alternative switches (OMB, 2003). Sensitivity plots are not only fundamental to determining which are the prominent input variables, but can be invaluable indicators of whether a particular project should be pursued (Koller, 1999). In (Saltelli et al, 2004) sensitivity analysis have been presented as: "*those techniques will answer questions of the type 'which of the uncertain input factors is more important in determining the uncertainty in the output of interest?, or, if we could eliminate the uncertainty in one of the input factors, which factor should we choose to reduce the most the variance of the output?".* Sensitivity analysis is considered by some as a prerequisite for model building in an any setting, be it diagnostic or prognostic, and in any field where models are used (Saltelli et al, 2004). Kolb quoted in (Rabitz 1989, as cited in Saltelli et al, 2004) noted that theoretical methods are sufficiently advanced, so that it is intellectually dishonest to perform modeling with sensitivity analysis. In Oreskes et al (1994) it has been shown that sensitivity analysis is not treated as a tool to build or improve a model, but it represents one of the possible licit uses that can be done of the model itself. Chapman & Ward (2004) have defined "*risk efficiency*" as the minimum risk level for a given level of expected performance.

The principal output reports provides by SimLab® are presented in Figure 2 through Figure 7 (probability distributions assigned to model input parameters), Figure 8 through Figure 13 (histograms of the output value-Razem (Total), Figure 14 (uncertainty analysis of the output value-Razem (Total), Figure 15 and Figure 16 (sensitivity analysis based on the *Standardised* 

Projects involve risk by nature (Zwikael & Ahn, 2011). Reducing the level of risk is extremely important in projects, and indeed results of this study suggest that project managers often use risk management planning practices, consistent with previous studies (Kerzner, 2009; Ahmed & Kayis, 2007; Voetsch, 2004; Zwikael, 2004, as cited in Zwikael & Ahn. 2011). The Project Management Body of Knowledge (PMBOK) defines risk management as one of nine project knowledge areas, alongside other topics such as scope, schedule, quality, and cost management (PMI Standards Committee [PMI], 2008). In some project contexts, risk management is perceived as a separate activity (Zwikael & Ahn,

 In countries with low levels of uncertainty avoidance, project managers place lower importance on risk management and hence do not always follow required processes. In industries with low levels of maturity, project managers do not frequently perform

When applying for European Union (EU) funding of projects, cost-benefit analysis (CBA) practitioners need to prepare comprehensive investment appraisals following the latest guidelines on CBA provided by the European Commission (2008). Since this Guide includes the need to conduct a proper risk analysis, partly through sensitivity analysis (Evans &

The purpose of sensitivity analysis is to determine the relationships between the uncertainty in the independent variables used in an analysis and the uncertainty in the resultant dependent variables. Sensitivity refers to the amount of uncertainty in a forecast that is caused by the uncertainty of an assumption as well as by the model itself. Sensitivity analysis can be used to find "*switch points*" -- critical parameter values at which estimated net benefits change sign or the low cost alternative switches (OMB, 2003). Sensitivity plots are not only fundamental to determining which are the prominent input variables, but can be invaluable indicators of whether a particular project should be pursued (Koller, 1999). In (Saltelli et al, 2004) sensitivity analysis have been presented as: "*those techniques will answer questions of the type 'which of the uncertain input factors is more important in determining the uncertainty in the output of interest?, or, if we could eliminate the uncertainty in one of the input factors, which factor should we choose to reduce the most the variance of the output?".* Sensitivity analysis is considered by some as a prerequisite for model building in an any setting, be it diagnostic or prognostic, and in any field where models are used (Saltelli et al, 2004). Kolb quoted in (Rabitz 1989, as cited in Saltelli et al, 2004) noted that theoretical methods are sufficiently advanced, so that it is intellectually dishonest to perform modeling with sensitivity analysis. In Oreskes et al (1994) it has been shown that sensitivity analysis is not treated as a tool to build or improve a model, but it represents one of the possible licit uses that can be done of the model itself. Chapman & Ward (2004) have defined "*risk efficiency*" as the minimum risk level for a given level of

The principal output reports provides by SimLab® are presented in Figure 2 through Figure 7 (probability distributions assigned to model input parameters), Figure 8 through Figure 13 (histograms of the output value-Razem (Total), Figure 14 (uncertainty analysis of the output value-Razem (Total), Figure 15 and Figure 16 (sensitivity analysis based on the *Standardised* 

2011).

Kula, 2011).

expected performance.

the risk management process.

*Regression Coefficients*, SRC) and Figure 17 (sensitivity analysis - *Cobwebs plot* based on the *Standardised Regression Coefficients*, SRC) (Bieda, 2011). Based on the economic feasibility model presented in (Liberman, 2003), in this study used uniform distributions. Figure 15 and Figure 16 shown the results from the sensitivity analysis. The performance of the SRC is shown to be extremely satisfactory when the model output varies linearly or at least monotonically with each independent variable. MC analysis-simulation is the only acceptable approach for U.S. Environmental Protection Agency (EPA) risk assessments (Smith, 2006). Because all of the parameters of the economic model are independent, the using of the SRC is shown to be extremely satisfactory (Bieda, 2010).

SimLab® is didactical software designed for global uncertainty and sensitivity analysis, developed by the Joint Research Centre of the European Commission and downloadable for free at: http://simlab.jrc.ec.europa.eu (Simlab*,* 2004). The sampling techniques available in SimLab® are FAST, Extended FAST, Fixed sampling, Latin Hypercube, replicated Latin Hypercube, Morris, Quasi-random LpTau, Random and Sobol (Saltelli et al, 2004). SimLab® can also run models built in Microsoft Excel®. Using the SimLab® in order to determine the most relevant parameters sampling presented by sensitivity analysis, after selected the Monte Carlo (MC) simulation, sampling method, the optimal number of executions is considered at least 10,000 times.

There are available various commercial software packages in order to conduct the risk analysis using MC simulation. Among them, Risk® and Crystal Ball®, developed by Palisade Corporation and Decisioneering, respectively. Risk® was originally designed for business application and is easy to use without a need for extensive statistical knowledge and modeling capacity (Sonnemann et al, 2004). Crystal Ball® is a simulation program that helps analyze the uncertainties associated with Microsoft Excel® spreadsheet models by MC simulation (Sonnemann et al, 2004). Another function of the Crystal Ball® is the sensitivity analysis (Bieda, 2007).

According to Hullet (2004) the estimate of project cost risk can be made more accurate and better understood if the sources of risk are disaggregated into those that affect time and those that affect the burn tate per unit time. The schedule risk and cost risk analysis have been conducted in Microsoft Excel® and Crystal Ball®. In conclusion, Hullet (2004) muses that cost risk analysis that explicitly incorporates schedule risk analysis results, merging them with burn rate risk information in the estimates of cost risk that are more accurate than the typical approach. In the opinion of Leach (2005) anyone who is serious about realistically forecasting project schedules, in other words, truly managing projects, rather than just monitoring them, should be using MC simulation software to plan and analyze projects stochastically. Stochastic simulation (often called MC simulation) allows to capture and understand the uncertainty inherent in the project. Anderson (2005) in paper presented to the Denver Crystal Ball Conference in 2005 discusses the results of the using the MC simulation instead the analytic approach in the nuclear power plant steam generator repair/replacement cost/benefit analysis (before nuclear power plant steam generator replacement decisions have never included a MC simulation) and the strengths of the weaknesses of using Crystal Ball® and MC simulation. In conclusion, Anderson (2005) seems to think that MC approach was clearly appropriate to fully assess the impact of any

Risk Analysis of the Waste to Energy Pyrolysis Facility Designs for City of Konin, in Poland, Using SimLab®

Fig. 3. Probability distributions assigned to input - Etap 2.

Fig. 4. Probability distributions assigned to input - Etap 3.

Toolpack 253

decision on the ratepayer. Selection of data play a key role in application of risk analysis to project investments. In most industries the costs of raw materials and component parts constitute the major cost of a product – in some cases up to 70 per cent (Azadi & Saen, 2011). In this study the most likely Total Project Cost values are about 2.53563E+007 USD and 2.663226E+007 USD for the analyzed Scenario. Every manager has a different degree of aversion to risk In this study the most likely Total Project Cost values are about 2.53563E+007 USD and 2.663226E+007 USD for the analyzed Scenario. Every manager has a different degree of aversion to risk.

Positive coefficients indicate that an increase in assumption is associated with an increase in the forecast, negative coefficients imply the reverse (Evans & Olson, 1998). In the Sensitivity Charts (Figures 16 and 17) is presented that variables Etap2 is the most influential parameter (98.93%) in the Project Total Costs.

Fig. 2. Probability distributions assigned to input - Etap 1.

decision on the ratepayer. Selection of data play a key role in application of risk analysis to project investments. In most industries the costs of raw materials and component parts constitute the major cost of a product – in some cases up to 70 per cent (Azadi & Saen, 2011). In this study the most likely Total Project Cost values are about 2.53563E+007 USD and 2.663226E+007 USD for the analyzed Scenario. Every manager has a different degree of aversion to risk In this study the most likely Total Project Cost values are about 2.53563E+007 USD and 2.663226E+007 USD for the analyzed Scenario. Every manager has a

Positive coefficients indicate that an increase in assumption is associated with an increase in the forecast, negative coefficients imply the reverse (Evans & Olson, 1998). In the Sensitivity Charts (Figures 16 and 17) is presented that variables Etap2 is the most influential parameter

different degree of aversion to risk.

(98.93%) in the Project Total Costs.

Fig. 2. Probability distributions assigned to input - Etap 1.


Fig. 3. Probability distributions assigned to input - Etap 2.


Fig. 4. Probability distributions assigned to input - Etap 3.

Risk Analysis of the Waste to Energy Pyrolysis Facility Designs for City of Konin, in Poland, Using SimLab®

Fig. 7. Probability distributions assigned to input - Etap 6.

Fig. 8. Histogram results for Etap 1.

Toolpack 255


Fig. 5. Probability distributions assigned to input - Etap 4.


Fig. 6. Probability distributions assigned to input - Etap 5.

Fig. 5. Probability distributions assigned to input - Etap 4.

Fig. 6. Probability distributions assigned to input - Etap 5.


Fig. 7. Probability distributions assigned to input - Etap 6.

Fig. 8. Histogram results for Etap 1.

Risk Analysis of the Waste to Energy Pyrolysis Facility Designs for City of Konin, in Poland, Using SimLab®

Fig. 11. Histogram results for Etap 4.

Fig. 12. Histogram results for Etap 5.

Toolpack 257


Fig. 9. Histogram results for Etap 2.

Fig. 10. Histogram results for Etap 3.

Fig. 9. Histogram results for Etap 2.

Fig. 10. Histogram results for Etap 3.


Fig. 11. Histogram results for Etap 4.

Fig. 12. Histogram results for Etap 5.

Risk Analysis of the Waste to Energy Pyrolysis Facility Designs for City of Konin, in Poland, Using SimLab®

Fig. 15. Sensitivity analysis tabulated value (SRC) for the 95% confidential level.

Fig. 16. Sensitivity analysis main Panel (SRC) for the 95% confidential level.

Toolpack 259


Fig. 13. Histograms results for Etap 6.

Fig. 14. SimLab® uncertainty analysis for RAZEM.

Fig. 13. Histograms results for Etap 6.

Fig. 14. SimLab® uncertainty analysis for RAZEM.


Fig. 15. Sensitivity analysis tabulated value (SRC) for the 95% confidential level.


Fig. 16. Sensitivity analysis main Panel (SRC) for the 95% confidential level.

Risk Analysis of the Waste to Energy Pyrolysis Facility Designs for City of Konin, in Poland, Using SimLab®

978-1-405-17517-3, Chichester

ISSN 0953-5543 (print)

Kraków, Poland (Polish)

Krakow

**7. Acknowledgments** 

**8. References** 

using the software package SimLab® for project risk management.

analyzed Scenario. Every manager has a different degree of aversion to risk.

This research scientific is granted by science financial support for 2011-2013.

Uncertainty reduction in the project is performed during the planning phase of the project

In summary, integrating risk analysis into waste to energy pyrolysis facility project management processes may be useful for the project managers. In this study the most likely Total Project Cost values are about 2.53563E+007 USD and 2.663226E+007 USD for the

Anderson TS. (2005). Diablo Canyon Power Plant: Steam generator repair/replacement

Azadi M, Saen RF. (2011). Developing a WPF-CCR model for selecting suppliers in the

Bieda B., Tadeusiewicz R. (2008). Decision support systems based on the Life Cycle Inventory

Bieda B. (2010). Decision Support Systems Based on the Economic Feasibility Assessment for

Bieda B. (2007). Assessing the Economic Feasibility of the Waste to Energy Facility Using

Chapman C, Ward S. (2004). Why risk efficiency is a key aspect of best practice projects.

European Commission (2008). *Guide to Cost-Benefits Analysis of Investment Projects*, Brussels,

Evans DJ, Kula E. (2011). Social Discount Rates an Welfare Weight for Public Investment

*Model Output, SAMO*, ISSN 1877-0428, Mediolan, October 2010

*International Journal of Project Management,* 22(8), pp. 619-631

Directorate General Regional Policy

no. 1, pp. 73-107, ISSN 0143-5671

*Transactions in Operational Research*. Vol. 15, Nr 1, January 2008, pp. 103-119 Bieda B. (2011). *Metoda Monte Carlo w Ocenie Niepewności w Stochastycznej Analizie Procesów* 

<http://www.anarisco.com.br/gerenciador/uploads/energia\_a2.pdf> Astrup T., Bilitewski B. (2010). Pyrolysis and Gasification, In*: Solid Waste Technology &* 

cost/benefit analysis. *Proceedings of the 2005 Crystal Ball User Conference,* Denver, Colorado, May 2005. Date of access 25 November 2011, Available from

*Management, Chapter 8.8,* Christensen, pp. 502-512, John Wiley & Sons, Ltd, ISBN

presence of stochastic data*., Operational Research Society*, Vol. 24, Nr. 1, pp. 31-48,

(LCI) for Municipal Solid Waste (MSW) Management under Uncertainty. *International* 

*Wytwórczych i Ekologii.* Wydawnictwo Naukowe AGH, ISBN 978-83-7464-344-3,

Municipal Solid Waste (MSW) Management Under Uncertainty Using Simlab® Toolpack. *Proceedings of the Sixth International Conference on Sensitivity Analysis of* 

Crystal Ball. *Proceedings of the 2007 Crystal Ball User Conference*., May 2007, Denver, Colorado, USA. Date of Access 25 November 2011, Available from < http://www.anarisco.com.br/gerenciador/uploads/telecomunicacoes\_1a.pdf> Bieda B. (2005). The role of thermal treatment in an integrated waste management, In: *Waste* 

*Recycling*, Wzorek, Kulczycka, Fečko, Kušnierowa, pp. 104-113, ISBN 83-89174-83-9,

Decisions under Budgetary Restrictions: The Case of Cyprus. *Fiscal Studies*, vol. 32,

Toolpack 261

Fig. 17. Sensitivity analysis (*Cobwebs plot*) (SCR - *Spearman Rank Correlation*). (SRC) for the 95% confidential level.

When the 10,000 trials are completed, the histograms provide by SimLab®, given in Figure 9 through Figure 14, present *statistics summary.* The "*Mean*", "*Variance*", "*Standard deviation*", "*Skewness*" and "*Kurtosis*" values form the basis of starting points for the analysis.
