CC BY
113. Grishnova, O. & Kozlovs'kyy, A. (2014). "Intelektual'nyy kapital Ukrayiny: intehral'na otsinka y porivnyal'nyy analiz". Ukrayina: aspekty pratsi, vol 1, 3-9.
4. Kaplan, R., & Norton, D., (1996). The Balanced Scorecard: Translating Strategy into Action, Harvard Business Schoo lP ress.
5. Lev., B. (2002). Intangibles: Management, Measurement and Reporting, Brooking Institution, Washington.
6. Luthy, D. (1998). Intellectual Capital and its Measurement, Asian Pacific Interdisciplinary Research Conference in Accounting (APIRA).
7. Nenkov, N. (2014). Identification of project ideas for practical applications of models and algorithms of artificial intelligence. Monograph. Edited by I. Spica. 6 [vol.]. Riga: Scientific institution Business competence Centre, 150 - 164.
8. Petrova, M. M., & Nenkov, N.V. (2015). Information models for assessing the maturity of the websites of Judicial authorizes in Bulgaria, Civitas, ISSN: 2217-4958, MMXV, N 9, Fakultet za pravne i poslovne studije, Novi Sad, 139-154.
9. Ruus, Y., Payk, S., & Fernstrem, L. (2008). Intellektualnyiy kapital: praktika upravleniya [Intellectual capital: Management Practices], Izd-vo «Vyisshaya shkola menedzhmenta», Saint Petersburg, Russian Federation.
10. Sergeev, A. L. (2005). "Tipologiya resursov intellektualnogo kapitala firmyi" [Typology of resources for company's intellectual capital], Ekonomicheskiy vestnik RGU, vol. 1, 59-64.
11. Stewart, T. (1991). Brainpower?, Fortune, June 3, 42-60.
12. Stewart, T. A. (1997). TheIntellectual Capital. The New Wealth of Organizations. N.-Y.-L., 270.
13. Tobin, James (1969). "A General Equilibrium Approach to Monetary Theory". Journal of Money, Credit and Banking 1 (1), 15-29.
14. Williams, M. (2000). Is a company's intellectual capital performance and intellectual capital disclosure practices related? Evidence from publicly listed companies from the FTSE 100, McMasters Intellectual Capital Conference.
15. Zinov, V. G. (2005). Menedzhment innovatsiy: kadrovoe obespepechenie: ucheb. posobie [Innovation management: staffing: textbook], Delo, Moscow, Russian Federation.
SIMULATION IN PROJECT RISK MANAGEMENT
PhD Mukhanova G. S., Phd Tymbayeva Zh. M.
Kazakhstan, Almaty, Kazakh National Research Technical University after K.Satpayev
Abstract. The article describes the Monte-Carlo simulation method for project risk management. The method includes a sensitivity analysis and simulation experiments. The simulation of parameters is performed with the MS Excel software and "EVA - risk analysis " software application on a sample investment project for an office building construction.
Key words: Project risk assessment, Monte-Carlo method, simulation modeling.
There are quantitative and qualitative methods of investment projects risk assessment. Qualitative assessment is a simpler though less accurate and reliable method while quantitative methods are used if the volume of information and indicators are sufficient to identify the likelihood or impact of risk [1]. In practice, both methods are used. The Monte Carlo simulation is a quantitative method.
Let us consider the use of the Monte Carlo method on a sample investment project "Construction of an administrative business building" by the "LuxGarant" LLP. The project involves construction of an office building to offer rent in the future for the recovery of project investments. The project cost is KzT290 247 000 with duration of nine months. The objective is to evaluate the risks of the project.
The Monte Carlo method is conducted in MS Excel environment using the "EVA - risk analysis" add-ons, a product of «EMF Plus". This software application is necessary for carrying out
simulation calculations of the investment project profitability, needed for the following reasons: implementation of repetitive simulation experiments, significant simplification of the simulations results processing and a graphical presentation of the results [2, 3].
In the first stage a sensitivity analysis is carried out. Figures 1 and 2 show the first steps of the sensitivity analysis by the EVA application - initial data input and parameters selection.
Fig. 1. Initial data input
Fig. 2. Parameters input
The initial data are accepted as follows:
- a calculated break-even point of the project, in the form of net present value (NPV), at the level of 107 million tenge per year;
- main components affecting the NPV - average amount of revenue, amount of current and general expenses and depreciation amount;
- other components: discount rate, tax rates on property, land and profit are constant.
The purpose of sensitivity analysis is to find the level of impact of the changes in the original data of a model (project) on the totals (profitability, revenue and/or payback period - any of the chosen options). The sensitivity analysis is to identify the critical factors of the boundaries change, for example, a maximum possible reduction of the volume of sales or prices for the products, works or
services for which the net present value (NPV) is positive. The wider the range of parameters in which performance indicators remain within acceptable levels, the greater the project safety margin, the better it is protected against fluctuations of various factors that influence upon the project results.
The sensitivity analysis algorithm is as follows:
- parameters of interest are chosen (automatically or manually);- the chosen parameters are changed alternately by reducing them for 10% and increasing
by the same amount, then the final figure is recalculated and shown on the "Tornado chart";
- if you change the parameters sequentially from -30% to + 30% in increments of 10%, the final schedule will be of the "spider" type.
Here are the two scenarios for analyzing the sensitivity of the project. In the first case, the range of the input parameters variation is -10% - +10%. In the second case, the range will be between -30% -+30%. By setting the required options/parameters, we obtain the following data shown on Figures 3 and 4.
Each bar shows how the final option changes when the corresponding input parameter is varied within its range (in this case, plus or minus 10 % and 30 %), and other input parameters remain equal to the base value. Longer bars correspond to the options with the greatest impact on the total and are always located at the top.
Susith p
Seiraoe KIHü
1191»»
.^«iiiïliiiiitof«p?rafiD|;iinl|>ei]friil nmiiicuta iKii:,« :kilîcs
I
DeprecuHmi UTSJIL! K?nW
maafiû —mu, innaji uhwi niuox voaifit iah fit
Summary analysis of paisible tarai values options
Ranîe Input option Cet Minimum Manmm Batvsitoi SsHMy
Total Option Total Op don OpDOU Total
Value Chans? (W) Vahle Value Change (K) Value
1 Revenue S3 96177 •10.67% 142191 119157 10.67% 1737S9 1Î7990 107667 229SÜ
Annual sum of opi ratine and general 94 HB4U 106921 4,69% 11393 10257 10766" 1J92
3 Depreciation 85 10777 0:08% 1113 107577 -4,08% 1361 1237 107667 ISO
Fig. 3. The sensitivity analysis results in EVA application. Scenario №1
Fig. 4. The sensitivity analysis results in EVA application. Scenario №2
The calculations and graph show that which exactly initial parameters change withstand the project, while remaining cost-effective. If this is a wide range, the sensitivity of the project can be considered low, and the business plan in this case is affordable.
From the analysis we conclude that the NPV> 0 for any variation of the input variable parameters such as the average amount of revenue, amount of current and general expenses and amount of depreciation in the range -10 % / 10%. At the same time, amount of revenue influences upon the NPV most of all.
The next step is the Monte Carlo simulation which normally involves the model parameters with the greatest impact on the result found in a sensitivity analysis. The Monte Carlo method algorithm is as the following:
- a set of variable parameters (risk factors) is selected;
- the changes boundary and the laws of distribution of each of the factors is put;
- the correlation matrix for each other dependent factors is clarified;
- calculations are performed for a desired number of experiments (generally not less than 1000).
In the course of the simulation, a large number of the model implementations is carried out under the joint random variation of selected parameters. The distribution law is found possibly based on historical data of a selected value. The user specifies the number of implementations of the method, and limits of changes with distribution law of a respective random variable for each of a selected parameter. If the chosen parameters depend on each other, it is necessary to set the correlation coefficient between them. Analysis by the Monte Carlo method enables us to understand the degree of uncertainty of the analyzed final value (characterized, for example, with specified 90% confidence interval). Also, the user can focus on the values of his interest, setting them instead of the probability in the determination of interval.
Figure 5 shows a first step in the Monte Carlo simulation which is the input parameters, that is, the possible risks. In this case, it is possible to reduce the volume of sales, the cost per square meter, the boundary changes are set with the confidence interval at 90%, the normal and triangular distribution laws being chosen for each of the risk factors.
A t £ D E | F | O | H | 1 1
EKpcnen ts
Kooiic 157J90 Mm* Cut* ¡-./[had
Annual sum ofopesilsi?; and fencraJ nnninE costs BB7 IllJ film ljiii.it opiian I ¡^»¡„rib.IiSnS^ii.t ! ■ ■ L'nrrv latiniL opriom '
Dejfecialktn 12J7 RmtU Addrfix evaT!njA<crl'E 1!; Nl"? ?Tfrr.t iin( l MUM
Taxable profit MfrJSS
]JK5TW BX »295
Net profit 117197 fr jrlllllllrj Bynlu* L«fl 5 ' M : *
Depf tciafliifl + 1 JS7 Ouipuf opnem. —w, SM:- | (SudiiiHiwi^i y^"iplrr(iftiit)ci Dftiaiii 3 | J UHtk ui xhrri
Net cast flmvs IIS 434
Discounted cash flow
Investment spending and iniejest oil the loan 0 Orlinax [iy-mmm* ■ Durjiion
11*434 Autrcij | X CueiI
Discenml i-w*tk ini 0,91
>'rl pKtinI vtluf XPV 107 6i 7
w 11
Fig. 5. Parameters input for the Monte Carlo simulation
Figures 6 and 7 shows the results of the software simulation for Scenario №1.
The table of results displays information in percentage, percentage probability that the value of the forecast will be less than or equal to a given percentile.
Next, let us try another scenario - how the NPV would be changed at the end of the payback period, if its components - the volume and cost of leased area will change within the confidence interval of 5% -95%. Let us consider a situation in which the maximum amount of leasable area is set at the level of 1281 square meters, the cost per square meter at USD 54.6$ with an even method of distribution. Figures 7 shows the results of the software simulation for Scenario №2.
ifi% 9 ÖS
It «2Л i 126 024,71
ч
V
if*. i " рч
имг;; JZfi 011,71
Sum тагу statistics
Minimum 69014 Left X 8S612
Maximum 145629 Left P 5,00%
Average 107555 Right X 125025
Standard deviation 11366 Right P 95.00%
Dlff X 37412,55
Oitt P 90%
/
/
/
/
Percentiles
m 88612,15 5054 1076SD,9S
ЮМ 92958,88 S5K 109044,80
m 95631,66 era 110593,22
20K 97844,27 65% 112032,40
25ÍS 99739,39 70% 113773,29
30% 101431,1 75% 115464,89
35K 103141,9 80% 117247,04
■»OSS 104929,9 8i% 119304,05
45K 106339,8 90% 121747,30
5096 107651 9J% 126024,71
Fig. 6. The results of the risk assessment by the Monte Carlo method.
Scenario №1
Fig. 7. The results of the risk assessment by the Monte Carlo method.
Scenario №1
Analysis by the Monte Carlo method shows that the minimal probability that the NPV of the project will be negative. This is possible only under a simultaneous deterioration of the risk factors. According to the distribution histogram, there is almost no likelihood that the NPV of the project would be negative.
Thus, assessment of the investment project risks for the construction of business buildings at different combinations of factors was conducted by using the Monte Carlo simulation method. As the research results show, the project is profitable, and the company's management is recommended to build this business building.
If the Monte Carlo method model is correct, it provides very reliable results, allowing insight into both the profitability of the project and its sustainability (sensitivity).
REFERENCES
1. Gazeev M.H. Indicators of the investments effectiveness in the market conditions. M., Economics, 2006. 125 p.
2. Tarasov I.V., Kotljarova E.N. Risk management of the investment project. Professional Magazine, 04 (65), April 2009. http://www.em.ru/er/2009-04/23714/ 05.11.2015.
3. Douglas W. Hubbard. How to measure anything. Assessment of intangible business. Trans. Pestereva E.M., Olymp-Business, 2009 - 298 p.
4. Mukhanova G., Tymbayeva Zh. Risk assessment in project management.//Project management development: practice and aspects. Materials of International Project Manager's Congress. -Astana, 2015. P. 89-94.
