Research Journal of Recent Sciences _________________________________________________ ISSN 2277-2502 Vol. 4(6), 53-61, June (2015) Res.J.Recent Sci. International Science Congress Association 53 Optimization of Bank Liquidity Management using Goal Programming and Fuzzy AHP Mohammadi R and Sherafati MNovin Pajoohan Research Institute, Department of Economics and Management, Tehran, IRANGraduate School of Management, Multimedia University, MALAYSIA Available online at: www.isca.in,www.isca.me Received 24st January 2014, revised 11th March 2014, accepted 10th June 2014Abstract In this research, Goal Programming (GP) and Fuzzy Analytic Hierarchy Process (FAHP) were integrated. The financial objectives of Parsian Bank (a leading private bank in Iran) were identified and prioritized. An optimal liquidity management model incorporating the following objectives was devised; capital sufficiency, liquidity risk, liquidity ratio, claims from other banks, investments portfolio, total consumption to total resources ratio, growth of total assets, fixed assets and other assets. Afterwards, the goal and structural limitations of variables were taken consideration and finally, the optimal liquidity management model was estimated. Then, by using the input variables (the liabilities side in the balance sheet and the related subsets) and the outputs (the assets side in the balance sheet and their subsets) in the period of 2011-2012, the optimal values for liquidity ratios and other items in the balance sheet were calculated using Lingo software. They were then compared with real values in the balance sheet accordingly. Next, the solutions and suggestions were offered for optimizing liquidity management in the bank. The eight objectives used for the preparation of optimal liquidity model were prioritized using the viewpoints of senior financial directors of the bank with the emphasis on a questionnaire, which was designed based on the FAHP method. Furthermore, in order to test the estimated model and assess its efficacy, the total return of the bank (R) was calculated once using the real items in the balance sheet, and another time using the values obtained from the model as well as the return formula (both for the period of 2011-2012). The results demonstrateda noticeable increase in the return of the estimated model in comparison with the real return of the bank. In addition, it should be stated that the estimated model can diminish the liquidity risk and increase the growth of the total assets of Parsian Bank, which evidently presents the reliability, validity, and application of the estimated liquidity management model for the banking system. Keywords: Optimization, liquidity management, ratios, fuzzy AHP, goal programming. IntroductionLiquidity risk is the risk due to lack of sufficient liquidity to cover short-term obligations and unexpected outflow of funds .This risk Includes both asset liquidity and funding liquidity risk. “Liquidity Management” is the ability of forecasting of bank’s liquidity in different time periods and financing these needs by minimizing the costs. Similar to other fields of management, liquidity management is based on the comparison between risk and productivity. Depositing more liquidity may result in decreasing the risk in the banks. However, the banks may lose the chance of investment and ended up to deduction of the productivity of resources. In addition, it is worthwhile to mention that smart liquidity management can help banks to reach the liquidity safe harbor and enable them to satisfy the liquidity needs of their depositor clients or the clients receiving facilities timely and with no error. Credit institutions usually convert their short-term liquid liabilities to long-term non-liquid assets. Although this conversion enables banks to protect their customers against liquidity problems, it also exposes the banks to liquidity risks. This unfavorable situation highlights the importance of liquidity management. In other words, liquidity management means to assure that the bank is capable of fulfilling its contractual obligations. In fact, liquidity management presents the bank’s ability to optimally manage the decreases in deposits and other debts, while managing the growth of loans portfolio, other assets, and other items out of the balance sheet. Consequently, the bank can compensate the deficiencies in its liquidity at an acceptable expense in the quickest possible time. This paper is an attempt to propose a model for optimizing liquidity management with GP perspective and integrating GP with FAHP. A variety of efforts have been undertaken to employ this model to estimate the optimal values of important liquidity ratios, the balance sheet items, and different items in financial statements of the bank. Review of Literature: Liquidity Optimization Models: In an optimization problem, we assume that all the parameters associated with the decision variables in the objective function or in constraint set are known and we need to find an optimal solution to it. An overview of the most important liquidity Research Journal of Recent Sciences ______________________________________________________________ISSN 2277-2502Vol. 4(6), 53-61, June (2015) Res.J.Recent Sci International Science Congress Association 54 optimization and measurement models in financial systems especially banks have been provided in the following: Baumol Maturity Model: Baumol used the economic order quantity in certain conditions. In uncertain situations where cash payment is high, this model may not be used and other models should be employed instead. Miller and Orr Model: This model controls the limits. To explain more, when the cash reaches its high limit, cash will be converted to negotiable instruments. On the other side, when the cash reaches the low limit, the conversion of negotiable instruments into cash will be set off. Therefore, when the cash level is positioned between the two limits, no transfer will be made. Stone Model: This model considers the present cash position as well as the cash position in the coming days. Demand for Bank Money: This model presupposes that the cash (central bank instruments) and bank money (bank deposits) cannot be substituted. Thus, two demand functions (one for cash and one for bank money) should be defined10. Bank Profit Maximization with Reserve Maintenance Presupposition: This model presupposes that the bank is not involved in any optimization challenge other than the profit modeling. According to this model, the bank can eliminate uncertainties and provide certain conditions10. Profit Maximization for Definite Capital Values: The aim of this model is to maximize profit with regard to the available capital along with the constraints imposed by policymakers10. Money Management Model: This is a framework for asset allocation. The goal function of this model is a geometric mean of holding period returns. Goal Programming in Capital Budgeting: This model is designed to determine and satisfy the goals of a firm in selecting capital projects. However, considering the special type of the model, a method for determination of liquidity from the inventory perspective has been proposed11. Related Research: In some studies, GP has been used alongside simulation analysis. In some other, Fuzzy GP has been utilized 12. In several studies, non-linier GP has been applied for managing assets and liabilities13. In some others, a combination of GP and AHP has been employed14. Table-1 Several previous related researchesResults Subject Authors and Year They show that volatility is the key driver behind (de-) centralization and provide an analytical solution for the 2- branch model and show that liquidity center can be interpreted as an option of immediate liquidity. Finally, they show that the n-branch model for real-world banking groups can be solved with high granularity within less than 30 seconds. Managing liquidity: Optimal degree of centralization Pokutta Schmaltz15 2010 They find that the patterns of insurance and hedging decisions are pole apart: cash-poor firms should hedge but not insure, whilst the opposite is true for cash rich firms. They also find non-monotonic effects of profitability. This may explain the mixed detections of empirical studies on corporate demand for hedging and insurance. Liquidity management and corporate demand for hedging and insurance Rochet Villeneuve16 2011 They found that banks reacted to the liquidity shock sensitively through an increase in their cash holdings not by liquidating bank loans but by selling securities in the financial market. Therewith, banks subjected to local financial contagion adjusted the liquidity of their portfolio mainly by actively selling and buying their securities in the financial market. Finally, there is no evidence to conclude that the existence of the lender of last resort mitigated the liquidity constraints in bank portfolio adjustments. Liquidity risk and bank portfolio management in a financial system without deposit insurance: Empirical evidence from Prewar Japan Sawada17 2010 Using a non-parametric method, we find that this prediction is supported by data, implying that banks effectively contain the disruption caused by the operational outage: payment flows between healthy banks remain unaffected. Banks’ intraday liquidity management during operational outages: Theory and evidence from the UK payment system Merrouche Schanz18 2010 They conclude that efforts to manage the liquidity crisis by banks led to a decline in credit supply. Liquidity risk management and credit supply in the financial crisis Cornett McNutt Strahan Tehranian19 2011 Research Journal of Recent Sciences ______________________________________________________________ISSN 2277-2502Vol. 4(6), 53-61, June (2015) Res.J.Recent Sci International Science Congress Association 55 Results Subject Authors and Year This Recent empirical research offered that access to lines of credit is contingent on the credit quality of the borrower as well as the financial condition of the lender. These findings suggest that lines of credit are an imperfect substitute for cash as a source of corporate liquidity. The use of bank lines of credit in corporate liquidity management: A review of empirical evidence Demiroglu James20 2011 They use a game theoretical framework to analyze the intraday behavior of banks with respect to settlement of interbank claims in a real-time gross settlement setting. Banks are risk neutral, but they show that most of the results are unaffected by risk aversion. The intraday liquidity management game Bech Garrarr21 2002 Overall, no study based on a combination of AHP and GP was found in the literature. Review of Fuzzy Numbers and Fuzzy AHP: Fuzzy numbers: Fuzzy numbers are in fact natural generalizations of ordinary numbers. An ordinary number like  can be shown with the following membership function:  \n \r \rTherefore, any real number can be stated as a fuzzy number. The simplest fuzzy numbers are triangular fuzzy numbers 22. We define a fuzzy number M on R to be a triangular fuzzy number, if its membership function  \n is equal to:                 !"#$%&\r'$The triangular fuzzy numbers can be expressed by (l, m, u). The parameters l, m, and u respectively indicate the smallest possible value, the most promising value, and the largest possible value that describe a fuzzy event. There are various operations on triangular fuzzy numbers. But here, the two important operations used in this study are illustrated. If we dene, two positive triangular fuzzy numbers (l, m, u) and (l, m, u), then:   ++ +  ,(-. \n( \n( Fuzzy AHP: AHP is one of the well-known multivariate decision making methods invented by Saaty in 1970s. Indices may be qualitative or quantitative. AHP is based on pair wise comparisons in which decision-maker forms a hierarchical decision tree and defines its indicators and choices. Then, she makes some pair wise comparisons and determines the weight of each factor in comparison with rival ones23. The traditional AHP method is problematic because it uses the exact value to express the decision maker’s opinion in a comparison of alternatives24. AHP method is often criticized due to the application of unbalanced scale of judgments and its inability to appropriately handle the inherent uncertainty and imprecision in the pairwise comparison process25. To overcome all these shortcomings, FAHP was developed for solving the hierarchical problems. Decision-makers usually nd it more condent to give interval judgments rather than xed value judgments26. In this study, the extent FAHP is utilized which was originally introduced by Chang. Let X = {x1,x2,x3, ... , xn} an object set, and G={g1 , g2 , g3 , ... , gn} be a goal set. According to the procedure of Chang’s extent analysis, each object is taken and extent analysis for each objective is performed respectively. Thus, extent analysis values for each object can be obtained, with the following signs: 120120124012  \r\n6748Where 12:\n64 all are Triangular fuzzy numbers. The steps of Chang’s extent analysis27 (can be given as follows: Step 1. With respect to the its object, the value of fuzzy synthetic extent is dened as: 0129=(+0129=(2=(,(To obtain ABC=(, the fuzzy addition operation of m extent analysis values for a particular matrix is performed as follows: 0129=(9=(EC=(FC=( And to obtain ??129=(9=(,(, the fuzzy addition operation of 12:\n674 values is performed such as Research Journal of Recent Sciences ______________________________________________________________ISSN 2277-2502Vol. 4(6), 53-61, June (2015) Res.J.Recent Sci International Science Congress Association 56 0129=(2=(G2=(EB=(FB=(I Step 2. As EF and EFare two triangular fuzzy numbers, the degree of possibility of EFK@EF is dened as K0    \n \rK \rK    !"#$%&\r'$Figure-1 illustrates equation-10 where d is the ordinate of the highest intersection point D between µMl and µM1. To compare M1 and M2, we need both the values of V(M1 M2 ) and V(M2 M1 )26. Step-3. The degree possibility for a convex fuzzy number to be greater than k convex fuzzy Mi (i = 1,2, ..., k) numbers can be dened by: V(M M1, M2, …, MK)=V[(M M1) and (M M2) and (M MK)]= Min V(M Mi) , i= 1,2,3,…, k Assume that d(Ai) = min V(Si Sk ) for k =1, 2, ... ,n; k i. Then the weight vector is given by NN4+NOwhere A i (i =1, 2, ... , n) are n elements. Step 4. Via normalization, the normalized weight vectors are: MNN4+NOwhere W is a non-fuzzy number. Research Methodology In this research, GP is integrated with multi-criteria fuzzy decision-making methods in order to optimize liquidity management. First, all the inputs and outputs of the cash management system were identified and determined. Then, the bank’s resources, consumption plans, and financial statements (including the balance sheet, profit and loss account, and cash flow statements) were reviewed. Moreover, important financial ratios affecting liquidity system were identified with regard to the experiences of other banks and the opinions obtained from bank financial experts. Subsequently, they were prioritized based on their importance using FAHP. All the liquidity system inputs and outputs were identified using the bank’s resources and consumption tables. Then, the goal and systemic limitations of the model were defined. Interviews with senior financial managers of banks showed 7 overall objectives for optimization of liquidity management. These objectives were presented to financial managers of Parsian Bank. Risk manager, financial manager, and their subordinates were interviewed and 8 specific objectives were extracted for Parsian Bank. Major objectives of Parsian Bank for liquidity management optimization included: capital sufficiency, liquidity risk, liquidity ratio, claims from other banks, investments portfolio, consumptions to resources ratio, total assets growth, fixed assets and other assets. Figure-1 The intersection between two triangular fuzzy numbers (Mand MThe above 8 objectives are not of equal importance for the bank. Therefore, by using AHP, the objectives are given coefficients. The outputs are determined in proportion with the importance of objectives. Suppose that the goal function of the GP for liquidity management is: 1 min m r n k krkr is the weight of objective and di is the positive or negative deviation from the objective. Therefore, a questionnaire was designed using AHP method and the weights of the 8 objectives were estimated. As mentioned above, all the objectives are not equally important for the bank and some of the objectives surpass the others in importance for the bank. Thus, all the objectives cannot have coefficient 1 and AHP method should be employed in order to optimally prioritize the 8 objectives. GP function calculates the outputs and optimal values for the balance sheet. Therefore, the most important objectives of the bank were estimated as the objectives of the bank can be different or even contradictory. Therefore, the two contrary objectives may not be met. Thus, for calculating the outputs, it should be recognized that which of the two contrary objectives will be estimated and this is possible through giving the weights to the objectives. Research Model: GP models consists of three parts. Decision variables (model inputs and outputs), objectives, and limitations. In this section, decision variables, objectives, and limitations of GP model of liquidity management in Parsian Bank are studied. Research Journal of Recent Sciences ______________________________________________________________ISSN 2277-2502Vol. 4(6), 53-61, June (2015) Res.J.Recent Sci International Science Congress Association 57 Identification of Objectives: In linear programming and GP models, the objective(s) of the decision-maker in terms of optimization are to be identified from the decision-makers point of view. Afterwards, the importance coefficient of each objective should be determined as GP estimates the objectives in sequence proportion with their importance. In other words, GP optimizes decision variables and model outputs, and hence, the most important objectives can be estimated in the first place. Sometimes some of the objectives may be contradictory and the model may fail in estimating them simultaneously. Thus, the most important ones were recognized. In this research, the objectives of the bank’s management for liquidity optimization were determined through several interviews. As already disclosed, the objectives include capital sufficiency, liquidity risk, liquidity ratio, claims from other banks, investments portfolio, total consumption to total resources ratio, growth of total assets, fixed assets and other assets. Model Decision Variables: Having devised the model’s objective function, decision variables and the model’s limitations should be determined. With regard to the balance sheet structure, the following definitions were declared for decision variables (the model’s input and output variables). Deviation from goal (d): Di in a GP model shows the limit in achieving goals. Since the principal goal of the GP model is to minimize the deviation, when the deviation is minimized, the goal achievement can be maximized accordingly. Thus, deviations are determined based on the limits with respect to the objective function. In this model, both positive and negative deviations are considered within the limits of the model, for any allocation of sums to any of the items. Therefore, it should be expressed that in the objective function, either the positive or the negative deviation or even both can be taken into account as the desirable goal. Using deviations in the objective function is dependent on the desires and interests of the designer or users of the model. If someone wants to reach a certain level of the goal (not less nor more), then one should take both positive and negative deviations. Thus, the second group of decision variables includes positive and negative deviations. Limitations: The required limitations for determination of the composition of the balance sheet items are presented in two groups: obligatory limitations (structural or systemic) and goal limitations. Obligatory limitations are stated in the form of limitations with higher and lower limits. Goal limitations are expressed with positive or negative deviations from the goal. The main objective function exhibits the deviation from the objectives and is dependent on the type of the objective, which defines the decrease in positive, negative, or both deviations. The model has 29 limitations, 21 of which are structural (related to the structure of the balance sheet) and 8 are goal limitations (related to the selected goals). Table-2 The first group of the model’s input, output, and decision variablesAssets Xi Liabilities and Equities Yi Cash X 1 Debts to Central Bank Y 1 Securities Debts to other banks and credit institutions Legal Reserve with Central Bank Ratio Demand deposits Deposits with other banks Investment deposits Y Short-term investments X 5 Other deposits Y 5 Credits granted and claims Dividends Y Claims and down payments Other payable accounts Long-term investments XTax reserve and severance compensation Tangible fixed assets X 9 Total equities Y 9 Intangible assets and other assets 10 - - Below-the-line items X 11 - - Obligations for guarantees 12 - - Source: The balance sheet and financial statements of Parsian Bank Table-3 The second group of model decision variablesDetermined goals Di + di - Capital sufficiency d 1 + d 1 - Liquidity risk d 2 + d 2 - Liquidity ratio d 3 + d 3 - Claims from other banks and credit institutions d 4 + d 4 - Investments portfolio d 5 + d 5 - Total consumption to total resources ratio d 6 + d 6 - Total growth of assets d 7 + d 7 - Fixed assets and other assets d 8 + d 8 - In addition, the model has 37 decision variables, including 21 principal variables (related to balance sheet items) and 16 deviation variables (8 positive deviations and 8 negative deviations). The limitations of the model were uncovered by using the interviews with senior bank managers added to the obligations prescribed by the central bank, and the relationships between the balance sheet and financial statements items with those of previous years. They are pointed numerically in the model presentation section. The model’s structural (systemic) limitations: The structural limitations are related to the ratios of the balance sheet items, Research Journal of Recent Sciences ______________________________________________________________ISSN 2277-2502Vol. 4(6), 53-61, June (2015) Res.J.Recent Sci International Science Congress Association 58 financial ratios, and liquidity that play significant roles in model estimation. These limitations were extracted from the obligations prescribed by the Central Bank, the structure of Parsian Bank, and the results obtained from the questionnaire and interviews. It is worthwhile to mention that the model has 22 structural limitations. In fact, the model’s target has been to obtain the optimal values for the assets-side variables (cash, short-term and long-term investments, securities, credits grant, etc.) with regard to the real values of the liabilities-side variables (such as the types of deposit). The values obtained from the model and software determine whether the credits that were granted lately, or the case that was maintained by the bank, or the amount of claims and down payments, or the amount of short-term and long-term investments, or the amount of fixed assets and other assets, in the balance sheet are optimal or not; and if not, how far they are away from the optimal values. Overall, the final return of the bank is calculated in two forms. The real values of the balance sheet and the optimal values extracted from the model will be calculated and compared accordingly. The final return will be a helpful benchmark to evaluate the efficacy of the model. Results and Discussion Model Estimation and Results: Results of FAHP: Based on the FAHP algorithm and with the aid of Excel software, the available options were prioritized as follow: First priority Liquidity risk with a coefficient of 0.288 Second priority Capital sufficiency with a coefficient of 0.243 Third priority Liquidity ratio with a coefficient of 0.190 Fourth priority Total consumptions to total resources ratio with a coefficient of 0.104 Fifth priority Total growth of assets with a coefficient of 0.062 Sixth priority Investments portfolio with a coefficient of 0.049 Seventh priority Claims from other banks with a coefficient of 0.040 Eighth priority Fixed assets and other assets with a coefficient of 0.025 In this section, the overall GP model for achieving the research objectives is formed. First, the model’s limitations including the goal and structural limitations are presented in parametric and numerical forms. Then, the model’s objective function will be presented using the coefficients of ratios and objectives, which were identified to optimize liquidity management in parametric and numerical forms. At last, when the model is absolutely devised, the model components will be entered in Lingo software and the optimized ratios and outputs will be achieved. GP Model Estimation: The final model is as shown here: 0\r8Q+66RNS +6T7NS+\nUNS+\nTNS+X6N+TUNS+TNNVS+6\NS.T: K+^_S__S_ Limitation of the bank’s cash reserve K+7_S__S_ Limitation of securities `+\n_S__S_+\n^_S_S_S__S_S_S_S_S_ Limitation of the statutory reserve ratio K+7_S__S_ Limitation of the investments or claims from other banks +6_S__S_ Limitation of short-term investments `+R_S__S_ Limitation of the credits granted and claims K+^_S__S_K+\n6 Limitation of claims and down payments +R_S__S_ Limitation of long-term investments K+\n_S__S_Limitation of tangible fixed assets and intangible assets (bK+X_S__S_ Limitation of other assets ((`+7S__S_S_(bS+\nLimitation of below-the-line items of the balance sheet +\nXS+6\S+6S+6S+6(*+\n^__+\n\URT\XT7+\nS__S_ Limitation of total return T76\n6R7 Limitation of debt to the Central Bank T^X\66 Limitation of debt to banks and other credit institutions ^67\n\n7U Limitation of demand deposits \n\\nX^^XTT Limitation of investment deposits 7\nX^ Limitation of other deposits +Rc Limitation of dividends ^X\n^ Limitation of other accounts payable X\n6T7\n Limitation of tax reserve \6UT\n Limitation of staff severance compensation (b\n6RU^\6 Limitation of equities Research Journal of Recent Sciences ______________________________________________________________ISSN 2277-2502Vol. 4(6), 53-61, June (2015) Res.J.Recent Sci International Science Congress Association 59 2=(+7^_SNSN9=3(b+RS+6SSSSSS(bS+6((SNSN2=(SS+\_SNSN9=(((2=(+^_SNN,SNNV(b9=((b2=(NSN\nU6677R^RS+7_(bSNN+R_S__S_SNNS(b+7_(bSNNThe values of variables obtained from the model are mentioned in the table-4 (retrieved form Lingo software): Table-4 Values of the variables obtained from the model and those entered in the balance sheet in 2011 (values of in million Rials)Variable Title Balance Sheet 2011 Model Results Different between the Model Values and Balance Sheet Values X1 Cash 4,065,625 3,209,661 -855,964 X2 Securities 55,000 103,878 +48,878 X3 Statutory reserve 24,487,164 28,321,098 +3,833,934 X4 Deposits with other banks 2,614,810 2,106,552 -508,258 X5 Short-term investments 162,454 229,668 +67,214 X6 Facilities granted 117,785,833 155,099,041 +37,313,208 X7 Claims and down payments 29,429,439 32,544,817 +3,115,378 X8 Long-term investments 2,307,459 2,665,913 +358,454 X9 Tangible fixed assets 3,024,267 2,870,437 -153,830 X10 Intangible assets and other assets 8,301,827 8,761,891 +460,064 Total assets 192,233,878 235,912,956 +43,679,078 X12 Securities 15,009,450 23,046,772 +8,037,322 Y1 Debt to the Central Bank 4,321,283 4,321,283 0 Y2 Debt to banks and institutions 4,706,522 4,706,522 0 Y3 Assets deposits 7,231,139 7,231,139 0 Y4 Investment deposits 151,677,644 151,677,644 0 Y5 Other deposits 3,100,670 3,100,670 0 Y6 Dividends 32,496 41,095 +8,599 Y7 Other accounts payable 7,601,700 7,601,700 0 Y8 Benefits reserve 612,431 612,431 0 Y9 Severance compensation reserve 52,941 52,941 0 Y10 Equities 12,897,052 12,897,052 0 Total of debts and rights 192,233,878 192,242,477 +8,599 Research Journal of Recent Sciences ______________________________________________________________ISSN 2277-2502Vol. 4(6), 53-61, June (2015) Res.J.Recent Sci International Science Congress Association 60 Discussion: In liquidity management, the fluid or cashable assets are more likely to be converted to cash promptly. These assets are the banks’ reserves to tackle any predictable and unpredictable economic fluctuation in balance sheet items. For example, when the financial markets are undeveloped and the claims can only be cashed on their maturity dates and there is no possibility for purchasing or selling them, it is suggested that the bank (especially in the bank under study) saves high amounts of cash in order to avoid paying the interest rates or prevent being paid less. In such conditions, the cashable assets can be approximately 10% (20% in critical situation) of the total assets of a bank. Expansion of markets throughout the world and diversity of financial tools have offered a financial flexibility in liquidity management for short-term plans, which in turn have decreased the need to maintain high amounts of cashable assets. In the developed financial markets, banks only keep 5% of their total assets as cashable assets. Thus, studying the markets before any kind of transaction is a must, because in some periods some assets seem to be cashable but it may be hard in practice to cash them at other times. The main objective of keeping cashable assets is to ensure that the predicted financial flows can meet the demands appropriately at the planned time. This obligatory investment can decrease the financial flexibility and increase the expense of granted credits for economic sectors. As credit expense increases, the bank’s financial risk level will increase consequently. In many countries, the growth of financial markets and the increase in investment portfolio generally reflex the growth of banks’ tendency towards innovative operations. In such situations investment portfolio includes different securities tools. This orientation in risk management means to benefit from the replacement of credit risk in the market prices fluctuations. As for the deposits, it can be said that deposits usually form a large portion of the bank’s liabilities. Customers’ deposits show the sums received from the public including savings, demand deposits, fixed deposits, deposits with former notification, and the deposits in foreign currencies. The structure and stability of deposits are the most important factors, which require serious considerations. The density, maturity, stability, and currency are the issues to be taken care of at the times it is needed to use these resources. The competition for equipping these resources is a normal issue in banks and most of the depositors including families and companies try to maximize the return of their sums. Thus, it is highly suggested that Parsian Bank deploys a viable strategy to absorb and maintain the deposits and adopt analytical procedures to study the orderliness, stability, and structure of deposits in order to use them in case it is needed to withdraw the resources. Another suggestion can be the provision of resources through inter-bank loans. When the bank finds it difficult to meet the demands for immediate cash from the depositors and supply financial resources, it may take an inter-bank loan. However, these resources are usually short-term and are used only in financial crises. Using such resources is a critical alarm and can decrease the validity of the bank. The expenses and amounts of inter-bank loans are strongly dependent on the validity of the bank. Moreover, frequent utilization of such resources can harm the reputation of the bank. The higher the reputation of the bank is, the lower will be the expense to use these resources and the higher can be the amounts of loans. Using such sources usually decreases the relative stability of plans. Therefore, it is suggested that the bank should be cautious when using this tool. The other suggestion is borrowing from the central bank. Borrowing from the central bank can be amongst the bank’s liabilities. The most important reason for borrowing from the central bank is to increase the volume of cash reserves as a result ofthe fluctuations in deposits. This change may occur when banks fail to predict their daily reserve status correctly and in order to fill the gap, they are forced to borrow. In such cases, the banks are helped in provisional resources supply conditions. Getting long-term credit from the central bank shows the abnormal situation of the bank; thus, borrowing from the central bank should be completely planned in order to evade the significant negative consequences. Conclusion Several important notes have been spotted in the above table (including the real values and the values obtained based on the bank’s objectives for liquidity management optimization), which can be summarized as follow: Firstly, the real values of the balance sheet for liabilities and assets are formed based on the natural trend and tangible and experiential obligations. In contrast, the values allocated by the model are based on the major objectives of the bank, the importance coefficients, and prioritization of objectives according to the legal and structural obligations in addition to the model limitations. Thus, it can be summed up that they obtained values are more scientific and by taking them into consideration, the bank’s managers can acquire useful insights to manage their liquidity efficiently. Moreover, in the estimated model, the majority of the liabilities side items in the balance sheet and those variables which values are out of the bank’s control (such as types of investments and equities) are estimated as input variables of the model. On the other side, based on the values for a specific year, the objectives, obligations, goal and structural limitations, and the optimal values for the assets side items, which are mostly under the bank’s control (such as cash, credits, claims from Central Bank and other banks, investments, fixed assets and other assets) are estimated as output variables. Secondly, the total assets in the estimated model displays 18% growth in comparison with the real values in the balance sheet. The main reason for this difference and the difference in resource allocation is that the model primarily tries not to exceed the determined limits in supply necessities and meet all Research Journal of Recent Sciences ______________________________________________________________ISSN 2277-2502Vol. 4(6), 53-61, June (2015) Res.J.Recent Sci International Science Congress Association 61 the objectives considering limitations and obligations. Thirdly, less sums are allocated to such assets that have low returns but are obligatory, compared with the real values. These items include cash, claims from other banks, fixed assets, other assets, etc. Furthermore, the items and assets with higher returns such as securities, credits, and investments are given more value than the real values. Another important note that can be expressed as a merit of the estimated model is that the bank’s model in the estimated form is higher than the real return of the bank and this can be definitely an asset for the bank to have. References 1.Barati Masoud, Dadashi Iman, Behzadnia Zahra and Zarei Samira, A Study of Risk Management in Iranian Banks, Research Journal of Recent Sciences,2(7), 1-7, (2013) 2.Bessis J, Risk Management in Banking, Wiley, (2004) 3.Moynihan G., Purushothaman P., Mcleod R. and Nichols W., DSSALM: A Decision Support System for Asset, Decision Support System, (2002)4.Soury D. and Vessal M., New methods of financing and liquidity management in banks, Proceedings nineteen Islamic Banking Conference, (2007)5.Milani A., Bank Liquidity Management Report, (2005) 6.Baumol William J, The Transactions Demand for Cash: An Inventory Theoretic Approach, Quarterly Journal of Economics, 66(4), 545-556(1952)7.Jain Sanjay and Arya Nitin, An Inverse Optimization Model for Linear Fractional Programming, Research Journal of Recent Sciences,2(4), 56-58 (2013)8.Miller M.H. and Orr. D., A model of the demand for money by firms, Quarterly Journal of Economics,80,413-435 (1966)9.Stone Bernell, The Use of Forecasts and Smoothing in Control-Limit Models for Cash Management, Financial Management, Spring, 72-84 (1972)10.Vince R., The New Money Management: A Framework for Asset Allocation, Wiley, New York (1995)11.Hawkins Clark A. and Adams Richard A., A Goal Programming Model For Capital Budgeting, Financial Management,3(1), 50-62 (1974)12.Arenas-Parra M., Bilbao-Terol A. and Rodriguez-Uria, M. V. A fuzzy goal programming approach to portfolio selection, European Journal of Operational Research,133, 287–297 (2001)13.Gordon H., Dash Jr. and Nina Kajiji, Nonlinear hierarchical modeling for efficient asset-liability management of property-liability insurers, (2004)14.Karimi Mahnaz, The optimal management of assets and liabilities of banks using the GP and AHP, Case Study: karafarin bank, (2007)15.Pokutta S and Schmalt Ch, Managing liquidity: Optimal degree of centralization, Journal of Banking and Finance,35, 627–638 (2011)16.Rochet J, Villeneuve S, Liquidity management and corporate demand for hedging and insurance, J. Finan. Intermediation, 20, 303–323 (2011)17.Sawada, Liquidity risk and bank portfolio management in a financial system without deposit insurance: Empirical evidence from prewar Japan, International Review of Economics and Finance, 19, 392–406 (2010)18.Merrouche ouarda and Schanz Jochen, Banks’ intraday liquidity management during operational outages: Theory and evidence from the UK payment system, Journal of Banking and Finance,34, 314–323 (2010)19.Cornett M and Mcnutt J, Liquidity risk management and credit supply in the financial crisis, Journal of Financial Economics, 101, 297–312 (2011)20.Demiroglu Cem and Christopher James, The use of bank lines of credit in corporate liquidity management: A review of empirical evidence, Journal of Banking and Finance,35(4), 775-782 (2011)21.Bech M and Garratt R, The intraday liquidity management game, Journal of Economic Theory, 109, 198–219 (2003)22.Jafari samimi A., Bidabad B. and Mohammadi R., Simulation of Continuous Qualitative Variables in Econometric Models Using Fuzzy Functions and Numbers, Australian Journal of Basic and Applied Sciences,4(10), 4780-4791 (2010)23.Saaty T.L., The analytic hierarchy process, New York: McGraw-Hill (1980)24.Deng H., Multicriteria analysis with fuzzy pair-wise comparison, International Journal of Approximate Reasoning,21, 215–231 (1999)25.Wang T.C. and Chen Y.H., Applying consistent fuzzy preference relations to partnership selection, Omega, the International Journal of Management Science,35, 384–388 (2007)26.Kahraman C., Cebeci U. and Ulukan Z., Multi-criteria supplier selection using fuzzy AHP, Logistics Information Management,16(6), 382–394 (2003)27.Chang D.Y., Applications of the extent analysis method on fuzzy AHP, European Journal of Operational Research,95, 649–655 (1996)