文档库 最新最全的文档下载
当前位置:文档库 › 软件代理模型A Proposed Security Model for Software Agent Security using DPHE and PKC

软件代理模型A Proposed Security Model for Software Agent Security using DPHE and PKC

A Proposed Security Model for Software Agent Security using DPHE and

Publiccryptosystem

Sachin Upadhye1, P.G. Khot2

Assistant Professor, Computer Application Deptt., Shri Ramdeobaba College of Engg. and Management, Nagpur

Email : sachupadhye@https://www.wendangku.net/doc/9213599294.html,

Professor, P.G.T.D., Deptt. Of Statistics, RTM Nagpur University, Nagpur

Email : pgkhot@https://www.wendangku.net/doc/9213599294.html,

Abstract--Software agent technology has become a driving force for recent advances in distributed systems. The concept of mobility in software agent is executable code, it raises major security problems. In this paper we deal with the protection of software agent code from possibly malicious hosts. We conceptualize on the specific cryptographic problems posed by mobile code. We are able to provide a solution for some of these problems.We present techniques how to achieve “non–interactive evaluation with encrypted functions” in certain cases and give a complete solution for this problem in important instances. We further present a way how an agent might securely perform a cryptographic primitive, blind signing, in an untrusted execution environment. Our results are based on the use of homomorphic encryption schemes and function composition techniques use Pailliar Cryptosystem.A propose a software agent security model for secured communication between the agents and platform. Data confidentiality is ensured using Dynamic programming with Pailliar cryptosystem additive/Multiplicativehomomorphic encryption property. The scheme ensures that the data possessed by theagents is secured at all times when it is executing at any of theuntrusted hosts. Paper also explain how the Additive homomorphicproperty of Pailliar scheme, Blind signature to be integrated with our model.

Weanalyse security aspects of mobility from acryptographic point of view. In section I introduce some fundamental of software agent security as well as discuss the constraints on cryptographic solution. Section II describing thePreliminariesof model such as Blind Signature, Pailliar Cryptosystem, Dynamic programming Homomorphic Encryption. Section III explain propose security model. Section IV focus on Discussion of the proposed model and Section V conclude the topic.

Keywords--Software agent Security, Dynamic programming, Homomrphic Encryption Techniques, Blind Signature, Pailliar cryptosystem

__________________________________________________*****_________________________________________________

Introduction:

Agent is anything that can be viewed as perceiving its environment through sensors and acting upon that environment through effectors. According to the IBM the software agent is a software entity that carries out some set of operations on behalf of a user or another program with some degree of independence or autonomy, and in so doing, employs some knowledge. Within the context of our research, we thus define an agent as an entity that receives inputs from its environment, evaluates the conditions and performs autonomous actions, perceiving and acting through its own environment to achieve its objectives. In Software Agent, Mobile Agent is regarded as an important new networking technology which, however, suffers from considerable security problems. Because it is based on the execution of mobile programs on remote and possibly untrusted computers, many observers question its fitness e.g., for E-commerce.

Security Problems:

We briefly describe the software agent code concept for discussing related security threats. This also enables to identify the specific constraints that mobile code imposes on cryptographic solutions [5]. The most evident security concern for mobile software agents is host security: hosts must be protected from the effect of foreign code which in much respect resembles network worms or computer viruses. The other concern is the security of the software agents themselves: the software agent?s code and data is at the full understanding of the executing host. So far little research was done on protecting a mobile agent from malicious hosts: the main focus was on making the execution of mobile code efficient and safe for the host. This list also points to the problems of using mobile code in security sensitive applications. In the case of E-commerce, for example, a shopping agent could be “programed”by a malicious server so it forgets the best prices collected tofar. Furthermore, itwould be unwise to let themobile agentdigitally sign an order form because this implies that theagent carries the user?s private key. Eavesdropping at the agent?s data also permits stealing attacks either on electronic money or on other electronic credentials (passwords, capabilities). Similar threats with perhaps even more disastrous consequences can be produced for active networks and battlefield agents too. Thus, the challenge for cryptography is to find answers to the following problems: _Can a mobile agent protect itself against

tampering by a malicious host? (code and

execution integrity)

_Can a mobile agent remotely sign a document without disclosing the user?s private key?

(computing withsecrets in public) _\

Can a mobile agent conceal the program it wants to have executed? (code privacy)

As security concern the software agent security threats can be generally classified into threecategories:-

Disclosure of Information

An agent may pose as awell-known service provider and tries to claim entity of atrusted

agent and try to convince the other agent with

creditcard no, bank account information or other

privateinformation.

Denial of Service

Agents can distribute false or useless information to prevent other agents from

completing their task correctly and on time for

example repeatedly sending messages that is

spamming agents with messages with cause?s slow

performance of agents. Sometimes an agent

participating in a transaction or communication

never took place which can lead to serious disputes.

Unauthorized Access

An Agent can directly interfere with

another agent by accessing and modifying agents

data or code which inturnchanges the agent

behaviour.

In our discussion we focus mainly on these three types of security threats. In any mobile agent paradigm mainly four types of Security threats can arise namely

Agent attacking another Agent

Agent attacking another Agent Platform

Agent Platform attacking an Agent

Others(Agent or Agent Platform) attacking another Agent Platform.

Constraints on Cryptographic Solutions

Network of Untrusted Nodes:

Creating a network ofmutually trusted nodes alleviates many mobile agent protectionproblems: users can trust the executing computersnot to tamper with their agents. Cryptographic solutions should presume anetwork of untrusted nodes.

Non-Interactive Protocols:

Ideally a security solutionfor mobile agents does not rely on an interactive protocolbetween the agent and its originating site: otherwise trulydetached operations become difficult and result in a quitelimited form of task delegation. In the example of theshopping agent, the user Alice would like to go off-lineinstead of keeping in touch with the shopping mobile agentshe sent off. Cryptographic solutions therefore should conceiveprotocols requiring minimal interaction between anoriginator and its mobile agents.

Provable Mobile Code Security:

We expect that mobilecode based applications will not be used in securitysensitive fields if provably security cannot be provided.Making tampering of agents just “difficult” without beingable to prove that there are no simple attacks seems to be toovague of a solution, even for risk management cost/benefitarguments. Consider, for example, E-commerce transactions: Because an agent?s buying actions should becomelegally binding operations, customers as well as providers require security guarantees. The protection offered to mobileagents must therefore be provable and truly cryptographic by linking the difficulty to alter an agent or spy itout to mathematically hard problems.. Cryptographic solutions for themobile agent protection problem therefore are subject tothe following constraints: Mobile agents should be allowed to execute in untrustedhosts but still have guarantees for their

correctexecution.

Mobile agents should not require interactive

protocolswith their originator.

Protection mechanisms should be provably secure. Software agent technology suffers from two drawbacks. The first is the possibility of being attacked by malicious unknown mobile agents. The second is that, in transactions, service hosts may breach the fairness principle. We, therefore, propose an improvement model that would not only protect service hosts but also guarantee the fairness of on-line transactions.

Preliminaries

Blind Signature

[10]proposed blind signatures for untraceable payments based on the RSA public-key cryptographic system[11]. Themajor contribution of Chaum?s scheme is the design of a signature protocol that allows signers toauthenticate applicants? identities, generate validdigital signatures for legal applicants, and verifythe signatures, but leave no clues as to whose signaturesthey are.In 1982, David Chaum invented a blind signature [12], that scheme allows the sender to have a given message signed by the signers, without revealing any information about the message or its signature. In 1996, Mambo, Usudu and Okamoto [13] proposed a new concept, proxy signature. In a proxy signature scheme, the original signer delegates his signing capacity to a proxy signer who can sign a message submitted on behalf of the original singer. Mambo, Usudu and Okamoto [14] proposed complete proxy signature, partial proxy signature and signature with an entitlement certificate. Zhang [15], and Kim, Park, and Won [8] proposed threshold proxy signature. In 2007, Li et al.[16] proposed a proxy blind signature scheme using verifiable self-certified public key, and compared the efficiency with Tan et al. In 2008 Xuang Yang and Zhaoping Yu proposed new scheme [17] and showed their scheme is more efficient than Li et al.[16] which is again modified by AungNway and NilarThein in 2009[18] and shown that their scheme is more efficient with low computation. This paper shows the scheme is more efficient and takes very less computational cost than the previous one. We study the security of blind signatures, especially for their application in electronic cash systems: we first define adequate security notions for blind signatures, then we propose the first schemes for which security arguments can be given.blind signature scheme is useful in several applications such as e-voting, e-payment and mobile agent environments. The security properties for a good proxy blind signature scheme are shows in Figure 1

Figure 1: Blind Signature Scheme property

Distinguishability:The blind signature must be distinguishable from the normal signature.

Non-repudiation:The original signer can sign message instead of the other party, the original

signer cannot deny their signatures against anyone.

Verifiability: The blind signature can be verified by everyone.

Unforgeability: Only the originatorsigner can create the proxy blind signature.

Identifiability: Anyone can determine the identity of the corresponding blind signature.

Prevention of misuse: It should be confident that key pair should be used only for creating Blind

signature, which conforms to delegation

information.

Unlinkability: When the signature is verified, the signer knows neither the message nor the signature

associated with the signature scheme.

To describe Chaum?s idea clearly, we delineate the concept of Chaum?s blind si gnature protocol using the following scenario. In this scenario, we assume that Bob wants to get a message M signed by Alice. Alice chooses two prime numbers, p and q. Let n =p*q. Let Alice?s private key be d and her public key be e. Chaum?s algorithm is i llustrated as follows:

1. Bob chooses a random number r?Z as the blind factor, produces a message digest H(MT for message M by a hash function H(), and uses rto blind the message digest H(M) Then he uses Alice?s public key e to encrypt this message to produce M!. M!=H(M)r e(mod n).

2. Bob sends the message M!to Alice.

3. Alice gets the signature σ(M!)by using M!andher private key to perform the computation as

follows:σ(M!) = M!d mod n

4. Alice sends the signature σ(M!)To Bob. After receiving σ(M!)Bob uses the blind factor r, performs the following computation to ……unblind?? his received message, and then generates the message signatureσ(M)

σ(M) = σ(M) / r (mod n) = H(M)d* r/r = H(M)

When Bob sends the pair of messages M and σ(M)to Alice, Alice uses the hash function to get therelative digest H(M!) and uses Bob?s public key todecrypt the signature to get H(M)!. Finally, Alice compares H(M) with H(M)!. If the result is equal, it means that the signature σ(M) is related to the message M, andσ(M) is indeed generated by Alice. Paillier Cryptosystem and there Additive Homomorphic Property (AHP)

Pascal Paillier introduced his cryptosystem in the 1999 published paper "Public-Key Cryptosystems Based on Composite Degree Residuosity Classes" [2]. The proposed technique is based on composite residuosity classes, whose computation is believed to be computationally difficult. It is a probabilistic asymmetric algorithm for public key cryptography and inherits additive homomorphic properties. The Definition of Paillier's Cryptosystem

Pick two large primes p and q and let n = pq. Let λ denote the Carmichael function, that is, λ (n) = LCM(p - 1; q – 1). Pick random g ? Zn2 such that L(gλmod n2) is invertible modulo n (Where L(u) = u-1/n). n and g are public; p and q (or λ) are private. For plaintext x and resulting cipher text y, select a random r ? Z n*. Then,

E K (x, r ) = g m r n mod n2

D K (y) =[L(yλ mod n2) / L(gλ mod

n2)] *mod n

PaillierAlgorithm:

1.Select two large primes, p and q

Distinguish-

ability

Verifiability

Non-repudiation

Unforgeability

Identifiability

Prevention of

misuse Unlinkability

2.Calculate the product n=p * q, such that

gcd(n,Φ(n)) = 1, where Φ(n) is Euler Function.

3.Choose a random number g, where g has order

multiple of n or gcd(L(gλ mod n2 ) n) = 1 where

L(t)= (t- 1) / n and λ(n)=lcm(p -1 q- 1)

4.The public key is composed of (g, n), while the

private key is composed of (p,q,λ).

5.The Encryption of a message m< n is given by:

c=gmrn mod n2.

6.The Decryption of cipher text c is given

by:m=(L(gλ mod n2 )/L(gλmod n2 ) )mod n. Additive Homomorphic Property

The additive homomorphic property of Paillier encryption follows that

E(X ?Y ) = E(X) ⊕E(Y).

And the generalized form is

E(mi l

i=1)=E(mi)

l

i=1

The Paillier scheme is known to be additively homomorphic. What might seem confusing at first is the fact that the two group operations are different,namely the product of two cipher texts will decrypt to the sum of their plaintexts.In comparison to that, the product of two RSA cipher texts decrypt to the productof their plaintexts. Hence the Paillier scheme is additively homomorphic and RSAmultiplicatively.

The given cipher texts c i are valid encryptions of plaintexts m i, c i = Enc(m i) =g m i r n i mod n2. The following properties hold

C1= g m1x1n mod n2

C2= g m2x2n mod n2

C1 * C2= g m1x1n. g m2x2n mod n2= g m1+ m2(x1

x2)n mod n2

Table 1: The homomorphic property of the Paillier

Cryptosystem

This means that the encryption of the addition of two plaintexts M1 and M2 is exactly the multiplication of the associated ciphertextC1 and C2.

Dynamic Programming Homomorphic Encryption Techniques

Dynamic programming [19] was developed by R. Bellman during the late 1950?s. Dynamic programming is a powerful method that can be applied to various combinatorial optimization problems. Many planning and control problems involve a sequence of decisions that are made over time. The initial decision is followed by a second, the second by a third, and so on. The process continues perhaps infinitely. Because the word dynamic describes situations that occur over time and programming is a synonym for planning, the original definition of dynamic programming was "planning over time." Dynamic programming has been described as the most general of the optimization approaches because conceivably it can solve the broadest class of problems. In many instances, this promise is unfulfilled because of the attending computational requirements. Certain problems, however, are particularly adaptable to the model structure and lend themselves to efficient computational procedures; in cases involving discontinuous functions or discrete variables, dynamic programming may be the only practical solution methodology. An example application of this protocol is the combinatorial auction, where multiple servers can solve a winner determination problem, i.e., they can find the combination of bids so that the sum of the bidding prices is maximized. Although the servers can compute the optimal solution correctly, the information of the bids that are not part of the optimal solution is kept secret even from the servers [20]. DP model represents a sequential decision process rather than an algebraic statement of a problem. The two principal components of the dynamic programming model are the states and decisions. A state is like a snapshot of the situation at some point in time. It describes the developments in sufficient detail so that alternative courses of action starting from the current state can be evaluated. A decision is an action that causes the state to change in some predefined way. Thus a decision causes a movement from one state to another. The state transition equations govern the movement. A sequential decision process starts in some initial state and advances forward, continuing until some final state is reached. The alternating sequence of states and decisions describes a path through the state space.

A proposed Security Model

Based on Constraints of cryptographic solution and software agent security issues it is sufficient to use a very simple one, consisting of only two main components: the agent and the agent platform. An agent comprises the code, state and data needed to carry out some computation. Multiple agents cooperate with one another to carry out some application. Mobility allows an agent to move or hop among agent platforms. The agent platform provides the computational environment in which an agent operates. An agent comprises the code, state and data needed to carry out some computation. Multiple agents cooperate with one another to carry out some application. Mobility allows an agent to move or hop among agent platforms. The agent platform provides the computational environment in which an agent operates. The platform where an agent originates is referred to as the home platform, and normally is the most trusted environment for an agent. It is assumed that the platform can eavesdrop on the agents data and communication hence confidentiality is required for both. It is also assumed that the platforms would not collude tocompromise the data. An agent platform may support multiple locations or meeting places where agents can interact. Figure 2, which show the movement of an agent among several agent platforms.

Figure 2: Software Agent Communication model

Software agent technology has been widely used to develop on-line shopping, auctioning, and other transactions. Users can set up software mobile agents and send them to collect product information, process an order, join an auction, pay for an order, and so on, instead of being totally involved in the transaction. Software agent technology enhances the convenience of electronic commerce. However, it suffers from two drawbacks. The first is the possibility that service hosts may be attacked by malicious unknown mobile agents. The second is that service hosts may breach the fairness principle in transactions. To overcome the first shortcoming, several solutions have been proposed. In our model we used Pailliar cryptosystem with Homomorphic additive property. The fairness principle is especially required for e-commerce money transaction, and other electronic commerce activities that emphasize the fair treatment of customers by service hosts. However, the current authentication scheme that is adopted by service hosts still has some fairness problem. It can be quite easily misused by unfair service hosts. Therefore, in this paper, we propose a fair and secure mobile agent environment. The proposed mobile agent based environment not only provides a novel authentication scheme that safeguards service hosts from being attacked especially malicious attack.

The commercial activity is a significant part of the network infrastructure allowing an open market of the services The commercial activity is a significant part of the network infrastructure allowing an open market of the services. Al Jaljouli et al [11] have implemented type of software agent in e-commerce to search and to filter information of interest from electronic markets. They describe also robust security techniques that ensure a sound security of information gathered throughout agent?s itinerary against various security attacks, as well as truncation attacks. The figure 6 describes the sequence of processes carried out during the agent?s lifetime. The author s utilize two co-operating agents where the initial verification terms are securely stored within a Software agent (SA) that resides at the initiator and cooperates with a Mobile agent (MA) that traverses the Internet. Nipur et al. [12] propose a fault tolerant comparison internet shopping system BestDeal. The author has conducted the simulation by launching nine shopping mobile agents where each has to visit five supplier sites to get the best deal for different products. Performance is measured in terms of execution steps as well as execution time of the simulation.. The design and implementation of a mobile agent platform for M-commerce applications is discussed in this paper. the advantage of adopting mobile agents for M-commerce is to scale up to large, dynamic world market places distributed over the Internet and to ease the access and participation of mobile users. We start by a discussion of the initial setup for the trading environment. In its simplest form, a trading environment is a market where some participants possess goods, and others want to consume them.

Steps of A proposed Model

1. A Software Agent host (SA) having the both the

Static and Dynamic Data, Static data is the Object

/ Task of the Agent which want to communicate

with Mobile Agent host (MA).

2.As SA is Blind Sign by the static data for the

authentication purpose and encrypted by the

private cryptosystem and SA use Pailliar public

cryptosystem to encrypt the dynamic data.

3.MA first authenticate the static data after that

decrypted by the static data and using public

cryptosystem of homomorhic encryption additive

/properly encrypted the MA resulted and pass it on

to next host

4.After completion of predefined path the SA

compute theauthenticate the static data and

decryption of dynamic data, and find out the

solution of Object / Task.

Software Agent

Figure 3. A proposed Model Software Agent Security model for E-Commerce

Conclusion:

Current E-Commerce buy/Sell product schemes employ an additive homomorphic encryption algorithm (e.g. Paillier encryption) to encrypt the sell value and exploit additive homomorphism of the encryption algorithm to recover the value for any host or choice with a single decryption. The contribution of this paper is a design of aadditivehomomorphicscheme. In additive homomorphicauction value, no single value isdecrypted in additivehomomorphic, so sell value for each host privacy is protected. The area of mobile agent security is still in somewhat immature state. Both the agent and the agent platform should to be protected by developing techniques and mechanisms.

Reference:

[1] T. Sander and C. Tschudin. Protecting Mobile Agents Against Malicious Hosts. In G. Vigna, editor, Mobile Agent Security, pages 44 60. Springer-Verlag: Heidelberg, Germany, 1998.

[2] Antonio Ma?a, Ernesto Pimentel, "An efficient software protection scheme", University of Málaga, Spain, 2002 [3] Joseph Tardo and Luis Valente, "Mobile Agent Security and Telescript," Proceedings of IEEE COMPCON '96, Santa Clara, California, pp. 58-63, February 1996, IEEE Computer Society Press.

[4] N. Jennings and M. Wooldridge, “Software Agents”, IEE Review, January 1996, pp. 17-20. [2] O.A. Ojesanmi and A. Crowther, “Security Issues in Mobile Agents”, International Journal of Agent Technologies and Systems, 2(4), pp. 39-55, October-December 2010, University, Nigeria.

[5] Upadhye-K hot, “Optimize Security solution for mobile agent security: A Review” International Journal Of Engineering And Computer Science, ISSN:2319-7242 Volume 2 Issue 1, Jan 2013 Page No. 322-329

[6] S. M..Moussa, G.A. Agha, “Integrating Encrypted Mobile Agents with Smart Spaces in a Multi-agent Simulator for Resource Management”, Journal of Software, Vol 5, No 6 (2010), 630-636, Jun 2010.

[7]Upadhye-khot“Study of Existing Security Enhancement for Software Agent” ,International Journal of Advance Research in Computer Science and Management Studies Volume 1, Issue 7, December 2013 ISSN: 2321-7782 (Online)

[8] Sakarkar-Shelke ―A New classification Scheme for Autonomous Software Agent‖, IAMA?09,2009, IEEE Int. Conf. , India

[9] S.M.S.I. Rizvi, Z. Sultana, B. Sun, and Md. W. Islam, “Security of Mobile Agent in Ad hoc Network using Threshold Cryptography”, World Academy of Science, Engineering and Technology 70- 2010.

[10] D. Chaum. Blind Signatures for Untraceable Payments. In Crypto ?82, pages 199–203. Plenum, New York, 1983. [11] R. Rivest, A. Shamir, and L. Adleman. A Method for Obtaining Digital Signatures and Public Key Cryptosystems. Communications of the ACM, 21(2):120–126, February 1978.

[12] D. Chaum, "Blind Signature Systems", Proceedings of Crypto’83, Plenum, pp.153. [2] B. Lee, H. Kim, and K.

MA

MA

Kim, “Strong proxy signature and its application”, Australasian Conference on Information Security andPrivacy(ACISP’2001), LNCS2119, Springer-Verlag, Sydney, 2001, pp.603- 608.

[13] M. Mambo& K. Usuda and E. Okamoto, "Proxy Signatures for delegating signing operation", Proc. 3rd ACM Conference on Computer andcommunications Security , ACM Press, 1996. pp.48-57.

[14] M. Mambo, K. Usuda and E. Okamoto, "Proxy signatures: Delegation of the power to sign messages", IEICE Trans. Fundamentals, 1996, Vol. E79-A, (9), pp.1338-1354.

[15] K. Zhang, "Threshold Proxy signature schemes", 1997 Information Security Workshop ,Janpan, 1997, pp.191-197.

[16] J.G. Li, and S. H. Wang, “New Efficient Proxy Blind Signature SchemeUsing Verifiable Self-certified Public Key”, International Journal of NetworkSecurity, Vol.4, No.2, 2007, pp.193–200.

[17] Xuan Yang, ZhaopingYu , “Efficient Proxy Blind Signature Schemebased on DLP”, International Conference on Embedded Software andSystems (ICESS2008).

[18] AungNwayOo and NilarThein, “DLP based Proxy Blind Signature Scheme with Low-Computation”, 2009 Fifth International Joint Conference on INC, IMS and IDC.

[19] E. Rasmusen. Games and Information.Blackwell, 1994.

[20] Makoto Yokoo, Koutarou Suzuki. Secure Multi-agent Dynamic programming based on Homomorphic encryption and its Application to Combinatorial Auctions.

基于代理模型方法的串列泵优化设计

第37卷第3期一一一一一一一一一一一哈一尔一滨一工一程一大一学一学一报一一一一一一一 一一一Vol.37?.3 2016年3月一一一一一一一一一一JournalofHarbinEngineeringUniversity一一一一一一一一一一一Mar.2016基于代理模型方法的串列泵优化设计 赵宇,王国玉,黄彪,王复峰 (北京理工大学流体机械工程研究所,北京100081) 摘一要:针对叶轮机械传统优化设计周期长二优化效率低等不足,提出了一种基于代理模型的优化设计方法,并应用于串列泵的优化设计,分析了关键设计参数对串列泵性能的影响三选择首二次级叶轮叶片安放角和相位角作为优化参数,选择效率和最小轴向截面平均压力系数作为优化设计目标函数,用来表示串列泵的外特性和空化特性三采用不同的方法建立代理模型,并采用敏感度分析方法和Paretofront方法进行参数影响分析和最优点的选取三采用数值计算的方法对优化后的串列泵外特性和空化特性进行验证,得到结论如下:提出的方法可以较好地应用于串列泵的优化设计,优化结果表明串列泵设计流量附近的效率和空化性能均得到提升三首二次级叶轮相位角对串列泵性能影响较小;首二次级叶轮叶片安放角对串列泵效率的敏感度之比和设计的载荷之比相同,首级叶轮叶片安放角是串列泵的空化特性的主要影响因素三 关键词:轴流泵;串列叶栅;系统优化设计;代理模型方法;空化特性doi:10.11990/jheu.201410085 网络出版地址:http://www.cnki.net/kcms/detail/23.1390.u.20160104.1648.020.html 中图分类号:TV131.32一文献标志码:A一文章编号:1006?7043(2016)03?438?11Optimizationdesignoftandempumpbasedonsurrogatemethod ZHAOYu,WANGGuoyu,HUANGBiao,WANGFufeng (ResearchInstituteofFluidMechanicalEngineering,BeijingInstituteofTechnology,Beijing100081,China)Abstract:Inthispaper,weproposeaglobaldesignoptimizationmethodbasedonasurrogatemodelforanaxial-flowtandempump,inordertoovercomethedisadvantagesofconventionaloptimizationmethods,includingtheirlongdesignperiodandlowerefficiency.Wediscusstheinfluencesoftheprimarydesignparametersonpumpper?formances,andusedfixedbladeanglesinboththefrontandrearimpellers,aswellasthephaseangle,asdesignvariables.Weselectedefficiencyandtheminimumaveragepressurecoefficientontheaxialsectionalsurfaceastheobjectivefunctions,whichrepresenttheenergyandcavitationperformances,respectively.Surrogatemodelswereconstructedbasedonvariousmethods.Inaddition,weusedglobalsensitivityanalysisandParetofrontmethodstofurtheranalyzethedesignparametersandoptimumpoint.Theresultsshowthattheoptimizationresultscanenhancetradeoffperformanceswell,withrespecttobothefficiencyandcavitationperformance,inthevicinityofthedesignflowofthetandempump.Theinfluenceofphaseangleonperformancecanbeneglectedrelativetotheinfluenceoftheothertwodesignvariables.Theimpactratioonefficiencyofafixedbladeangleinthetwoimpellersisthesameasintheirdesignedloadingdistributions.Afixedbladeangleinthefrontimpellerhasagreatinfluenceoncavitati?onperformance.Keywords:axial?flowpump;tandemcascades;globaldesignoptimization;surrogatemethod;cavitationperform?ances收稿日期:2014?10?31.网络出版日期:2016?01?04.基金项目:国家自然科学基金资助项目(51239005,51479002).作者简介:赵宇(1989?),男,博士研究生;王国玉(1961?),男,教授,博士生导师. 通信作者:王国玉,E?mail:wangguoyu@bit.edu.cn.一一串列叶栅是指多排叶栅直接相连的布置方式,具有灵活的气动特性,目前在航空发动 机[1?3]和水下螺旋桨推进中[4?5]已经得到应用三 串列泵是一种采用串列叶栅布置的泵,和普通的双级泵相比,由于减少一级导叶而显著减小泵级的轴向尺寸和质量,从而使得串列泵具有结构紧凑和功率密度高的特点;又由于两级动叶轮直接串联,前后级叶轮流动的相互作用会 导致前后级叶轮流场变化,因此串列泵具有特 殊的能量特性[6?7]三对于传统泵来说,效率和空化性能不可兼得[8],而串列泵通过调整前后叶栅的载荷分配可以兼顾效率和空化性能[9?10]三尽管串列泵具有上述诸多优点,但与传统泵相

委托代理理论论文

委托代理理论的主要内容及其在会计中的应用探析 摘要:委托代理理论是现代企业理论的重要组成部分,委托代理问题也是国有企业改革的热点问题。信息不对称和利益不一致将导致代理成本的产生,实证代理理论和委托人—代理人理论正是为研究如何减少代理成本而形成的两个不同的研究方向,有其各自的特点和方法。当然该理论也存在缺陷。 关键词:委托代理关系;实证代理理论;委托人—代理人理论 委托代理理论产生于20世纪60年代末70年代初,当时的经济学家们不满于阿罗———德布鲁体系中的企业“黑箱”理论,开始深入研究关于企业内部信息不对称和激励的问题,希望更全面理解企业这种经济组织,从而形成了“现代企业理论”。基于委托代理关系产生的委托代理理论是现代企业理论的重要组成部分。 1.委托代理关系的产生 委托代理关系是随着企业所有权和控制权(经营权)的逐步分离而产生的。在18世纪,业主型企业占主导地位,企业主即是经营者,所有权和经营权合二为一,不存在委托代理关系。到了19世纪中期,在生产社会化程度日益提高和市场联系扩大的过程中,企业规模不断扩大,这就要求经营者不但要具备经营管理知识,而且要有专业的技术知识,才能有效的组织企业进行生产,在市场竞争中获胜。在这种情况下企业经营者的来源和组成结构就发生了变化,一批新兴的职业经理人员逐渐形成,成为所有者的代理人旧时的“企业主企业”演化为现代的“经理人员的企业”。 现代股份公司这种企业组织形式对委托代理关系的发展起到了极大的推动作用。因为股东数量庞大且分散,不能对企业进行直接的管理和支配,必须把企业委托给经理经营,而经理就是所有者(股东)的代理人,拥有经营决策权。Bert和Me ans指出:现代公司的发展已经发生了“所有与控制”的分离,公司实际上已经为由职业经理组成的“控制者集团”所控制。Chandler则称之为“经理人员资本主义的兴起和企业主资本主义的衰落过程。”笔者认为,委托代理关系正是在所有权和控制权分离的基础上产生的。 2. 委托代理理论的内涵与主要内容 2.1委托代理理论内涵界定 所谓委托—代理关系,就是一种契约关系,通过这一契约,一个人或一些人(委

多任务委托代理模型下国企经理激励问题研究

多任务委托代理模型下国企经理激励问题研究① 袁江天1,2,张 维3,4 (1.南开大学经济学院,天津300071;21北方信托博士后工作站,天津300201; 31天津大学管理学院,天津300072;4.天津财经大学,天津300222) 摘要:运用多任务委托代理模型研究国企经理的最优激励合同后得出,若多工作任务努力的激励成本之间相互独立,那么激励相容条件下的各工作任务的最优业绩报酬也是相互独立的,且最优业绩报酬是绝对风险规避度、边际激励成本变化率和可观测变量方差的递减函数;而如果多工作任务激励成本是相互依存的,那么在激励相容条件下,政治性活动和满足上级偏好的最优激励合同为“门槛型激励合同”,即只有当国企经理所创造的业绩超过一定的“门槛值”时,对其的激励才是正向的,否则将是负向的.并且定量地给出了国企经理的“门槛型激励条件”. 关键词:国有企业;经理人激励;多任务委托代理模型;门槛型激励合同 中图分类号:F062.5 文献标识码:A 文章编号:1007-9807(2006)03-0045-09 0 引 言 关于企业经理激励问题最早的研究是T aussings和Baker在1925年完成的,但是与猜想不符的是,他们发现企业经营者报酬和企业业绩之间只有很小的相关关系[1].对此结果他们感到相当吃惊,于是呼吁加强研究以提示那些影响企业经理报酬的其他变量.在此之后,先后有委托代理理论、人力资本理论、劳动力市场理论、公司治理理论等,从不同角度对企业经理报酬与激励问题提出了自己的解释框架[2].居于主流的委托代理理论认为,企业所有者与经营者之间构成委托代理关系,由于两者之间信息不对称,为激励代理人为委托人努力工作,委托人应该以代理人工作绩效为依据向代理人支付报酬[3].但是传统的委托代理模型假定代理人只从事单一工作项目,代理人的努力选择也是一维的.模型证明了,如果委托人不能直接观测到代理的努力选择,代理人的报酬必须依赖于代理人努力下的经营业绩,因为只有这样,代理人才有积极性努力工作[4].企业理论认为,企业是一系列要素之间合同的纽结(a nexus of contracts)[5],并可简化为一个物质资本与人力资本(企业家才能和劳动力)的一种特殊契约[6].但对国有企业的经理来说,由于国有资本与国企经理之间的不对等契约关系剥夺了国企经理的剩余索取权,对国企经理的其中一个重要激励方式就表现为对企业的控制权激励[7,8].考虑到自己对企业的控制权来源于行政性的上级任命,而“如果失去了控制权,也就失去了一切”[9].公有制产权条件下控制权收益的不可补偿性和国企经理显性收入上边界的不可突破,造成了对国有企业控制权的争夺异常激烈.这样国企经理的行政性任命就使这场争夺演变为政治性争夺,而不是市场经济条件下普遍存在的企业控制权市场的争夺.这也就是“政企合谋”和国企经理的多任务委托代理的深层原因,导致国企经理在其岗位上所从事的工作绝不止经营性目标一项,而是至少包括了经营性目标、政治性目标、上级偏好等三个方面.经营性目标主要指企业价值(利润)最大化;政治性目标主要指维持“安定团结”、 “企业办社会”等特有的一些“社会责任”;上级偏好目标则主要指满足上级对自己工作的要求、维护上级的满意程度. 第9卷第3期2006年6月 管 理 科 学 学 报 JOURNA L OF M ANAGE ME NT SCIE NCES I N CHI NA V ol.9N o.3 Jun.2006 ①收稿日期:2005-10-10;修订日期:2006-04-25. 作者简介:袁江天(1970—),男,四川中江人,博士,北方信托与南开大学合作培养在站博士后.

委托代理理论案例

委托与代理理论 中国市场企业发展作为背景 、背景 农村孩子要想有出息,只有两条路:读书和打工。阿科选择了第 二条路,因为他的成绩不好,在这15 年里,他先是跟一个师傅学做木工3 年,出师后做了几年小木匠,再到江浙地区的家具厂做木工,后来在一个家具厂里学会了电脑绘图和设计欧式家具,并且从一个普通木工变成班长。去年年底,老板提拔他为经理,让他春节后招聘一批木工到厂里,同时设计一套管理体系。 从一个小学文化的小木匠,变身为管理十几个人的企业经理,阿 科的年收入大大增加了(应该比理论多),但是他也发现自己的知识远远不够了。企业理论与现实相结合,围绕企业的招聘、薪酬和管理等问题进行了多次对话。在理论和实践的交相辉映中,送给小木匠套有趣且有用的企业理论。 二、招聘 对话从招人开始。合格的员工能使管理事半功倍,因为只有好的 投入才有好的产出,更何况春节后是工人供给的高峰期。如果春节后招不到合适的工人,那么平时再去农村集中招人几乎不可能。

阿科:现在社会上很看重一个人的文凭。你说理论是不是应该尽可能招文凭高一点的,比如高中毕业生? 理论:未必。根据信号发射理论,文凭是显示能力的一种信号, 但未必适合于所有行业。你自己也是小学毕业,今天不是干得挺好吗?关键是,你们这个行业怎么判断一个人的专业水平呢? 阿科:哦,这个简单。就是问他有几年的木工经验。 理论:如果经验多的机会就多,那人家撒谎怎么办?比如,明明只干过一年木工活,但可能骗你说干了三年。你们这行也没有实习经 历证明啊。这样会导致你招到的人反而是没有足够工作经验的人,理论们把这种现象称为“逆向选择”。 阿科:但有一样东西是骗不了人的,就是工人手掌上的老茧。干活年份越久,肯定老茧越厚。同样是木工,拿刨子的和拿打孔机的又不同。而且木工的老茧长在手上的位置和其他行业的也不同。 理论:那就是理论们要的信号啊。如果是招干体力活的木工,那么看老茧肯定比看文凭更管用。好,现在理论们得到了企业理论的第一个原则。 原则一:看文凭不如看老茧

销售人员素质模型

从以下三个方面着手揭示了优秀销售人员所具备的特质: 1、由业绩差异发现优秀的销售特质: 与销售业绩相关性最大的前5项能力是:客户了解度、好胜好强、取悦心理、情绪稳定、大方自信。 与销售业绩相关性最小的因素是:文化程度、工作经验、年龄、工作时间。 2、由客户的满意度发现优秀的销售特质: 客户最喜欢销售人员的前5项因素是:能理解人、能拿出好主意、对人诚实可靠、对客户经常关心帮助、不怕被拒绝。 客户最讨厌的销售人员的表现是:话多、欺骗、不负责任、没主意、没耐性。 3、由行业特性和销售流程发现优秀的销售特质: 销售领域不存在通用型销售人才,一个知名品牌的销售人员则需要“耐心,悉心维护客户和一贯化的自我管理能力”。 在招聘过程中,需要主考官巧设问题,以此真实了解应聘者的特质。下面是对面试和场景测试给出的几项要点提示: ★面试 1、用统一标准提问。运用动机问题的提问方式代替无效的是非选择性问题。如“顾客反映产品价格偏高,你如何处理?” 2、避免光环效应。对应聘者的评分不要受其文凭、衣着、外貌或某人推荐等影响; 3、给应聘者以鼓励以期得到真实的答案。一般应聘者面对考官的提问,刚开始时会比较拘谨,有时甚至为了取悦面试官,会根据面试官脸上的表情来对自己的回答作出调整,这个时候,他所表达的也许包含了一些不太真实的东西。故而面试官应该用目光、用点头等形体语言,更多地给应聘者以鼓励,让应聘者放松下来,这个时候,得到的答案也许才是真实可靠的。 ★场景测试 场景测试应该是专为招聘销售人员而设置的。即设计销售场景或者让应聘者进入现实工作中一段时间。然后通过同事和客户的反馈,以期掌握应聘者的真实信息。当然,对于被考察者也有问题要答,比如,会让他写出“最讨厌的员工名单”等。一个合适的应聘人员列出的名单往往是极少,甚至是空白(一个有效的合作者看到别人更多的是优点),而缺乏亲和力的应聘者一般会写出一长串“讨厌者的名单”。显然,后一种人缺乏优秀销售人员应有的特质。

委托代理模型

委托代理模型与激励机制分析 委托代理模型分析的核心是解决代理人的道德风险和逆向选择问题,而这一问题产生的根源是由于委托和代理人信息不对称。激励机制的设计原理是,委托人通过一种分配制度来奖励代理人提供更多的信息,以缩小委托人和代理人之间的信息不对称。激励机制实质是对于不完全合同的有益补充,诱使代理人决策目标和委托人利益目标趋向一致。 一、委托代理模型的的基本假设 代理人的生产函数: ()ε+=a a y (1) 式中:y (a )是可以观测的产出结果;a 是代理人的行动;ε是白噪声,()2,0~σεN 。 现在假定代理人的报酬()y w 是产出y 的线性函数: ()y b s y w ?+= (2) 式中,s 是代理人的固定报酬部分,如工资;b 是代理人的浮动报酬部分,如奖金率或利润留成率。 代理人获得的收益1p 为: ()()a c y w p -=1 (3) 式中,()a c 是代理人采取行动a 的成本,由定义可知,()00=c ,即代理人 不采取任何行动时成本为0;这时,()[]()()00==+=εεE E a y E ,即E ().0=y () a c 具有性质,()0>'a c ,()0>''a c 。 委托人获得的预期收益2p 为: ()()y w a y p -=2 (4) 委托人获得的经期收益)(2p E 为:

()()()[]()[]()[]()()w E y E y w E a y E y w a y E p E -=-=-=2 (5) 二、风险态度中立的最优激励机制 经济学的一个基本假设是:每一个人都是以最少的付出获得效益最大化为根本目标的。这里假设委托人和代理人都是理性的,所以委托人的目标是企业价值的最大化,在这里具体为委托人获得收益1p 的最大化;而委托人和代理人获得最大收益的基础是企业可以创造财富P 的最大化,可以简化为21p p p +=的最大化,下面的分析假定委托人和代理人的风险态度都是中立的,即他们都是为了自身利益的最大化。 1、代理人最优目标的实现 委托人预期收入有一个最低下限标准,现在假设委托人最小收益目标是2p ,所以有:()()2p y w a y ≥-,为了让代理人的收益最大化,即在满足委托人最低收益2p 的前提下1p 尽可能的高,有()()2p v w a v =-。即()[]()[]()22p y E p a y E y w E -=-=,这时代理人的期望收益为: ()()()()()()()a c p a E a c p y E a c w E p E --=--=-=221 (6) ()1p E 获得最大值的一阶条件,即代理人实现最优目标的条件是:()02='p E ,即为()1='FB a C 。 3、委托人代理人共同获得最大收益目标的实现 委托人和代理人共同获得最大收益目标的实现,即实现21p p p +=的最大化。由(3)、(4)式可得: ()()()a c a a c a y p p p -+=-=+=ε21 他们联合的期望收益为: ()()()()a c a a c a E p E -=-+=ε (8) ()p E 获得最大值的一阶条件,即委托人实现最优目标的条件是:()0='p E ,即为()1='FB a C 。

供应链企业间的委托—代理关系.

供应链企业间的委托—代理关系 一.供应链企业间的委托—代理关系的提出及内容 最早提出委托代理(principaI-agent)概念的目的是为了研究股份制公司的管理体制问题。在股份制条件下,公司的所有权与经营权相分离,经理阶层代表股东行使管理职能。股东和经理层在利益上有时会不一致,但由于经理比股东更了解公司的信息,因此,经理可能做出与股东不相一致的行为。代理问题就是研究如何进行制度设计,使经理层在按自身利益最大化行动时,最大程度地实现股东的利益。委托代理问题是由当事人各方的信息不对称引起 的。信息不对称是指一方拥有另一方所没有的信息,拥有信息的一方称为代理方(agent), 缺乏信息的一方称为委托方(principaI)。信息不对称可以从时间和内容上划分。从发生的时间看,非对称可能发生在当事人签约之前(exante),也可能发生在签约之后(expOst);从内容上看,非对称可以是某些参与人的行动,也可能指某些参与人的信息或知识。因此,委托代理问题在经济生活中普遍存在,只要当事人各方在拥有的信息上具有不对等性,就存在着委托代理问题。政府和纳税人、公司和雇员、供方和需方等都存在代理问题。 二.应用实例 在市场交换中,存在着许多由于供应商和顾客的信息不对称,而引起的委托代理问题。NeIsOn研究认为,产品和服务具有经验属性(experienceattributes),其质量只有在购买之后才能发现。由

于顾客通常不能正确地辨别产品的质量,可能产生两个问题:第一,供应商不具备提供某种质量产品或服务的能力,而做出错误的质量承诺,而顾客又难以正确辨认供应商的能力,由此产生了有害选择(adverseseIectiOn)问题;第二,供应商可能在签约后采取欺骗行为,结果产生道德风险(mOraIhazard)问题。有害选择问题通常可以采用信号理论的方法解决,即利用某种信号揭示参与者的私有信息。例如,在汽车修理市场上,供应商通过某些不可回收(nOn-saIvageabIe)的投资(如设置标志等),显示其拥有较高的质量。因为低质量的供应商不愿意进行这方面的投资,通过这种方式就可以把高质量的供应商和低质量的供应商区分开来。道德风险问题则通过采用激励机制,如给予价格补偿,约束供应商的欺骗行为。三.供应链企业间的委托代理问题的特征 (l)供应链的企业间是一种合作竞争的关系 供应链的本质强调处于供应链上的企业间合作,强调企业集中资源发展其核心业务和核心竞争力,而对非核心业务则通过外包方式完成。供应链的思想与传统企业模式的根本不同在于,它改变了对产品链上其它企业的看法,供应链企业不再把它们看作是竞争对手,而是当作合作伙伴,为实现最终顾客满意的目标而进行协同生产,生产活动按整个供应链实行优化,而不是仅仅考虑本企业的利益。供应链企业间虽然强调合作,但也存在利益冲突,企业之间为分配合作带来的利益会展开竞争。因此,供应链企业的基础和目标是合作,但是由于它们利益主体的不同也存在竞争。研究供应链企业间的委托代理关系,就

委托—代理模型及其应用

委托—代理模型及其应用 摘要:本文将委托——代理理论模型归纳为采用激励和监督混合的有不确定性但可监督的委托——代理模型,采用递推归纳法及子博弈精炼纳什均衡分析这一模型。 关键词:委托-代理现代企业激励和约束机制博弈论信息经济学 1、委托代理理论的主要观点 委托——代理理论的主要观点认为:委托——代理关系是随着生产力大发展和规模化大生产的出现而产生的。其原因一方面是生产力发展使得分工进一步细化,权利的所有者由于知识、能力和精力的原因不能行使所有的权利了;另一方面专业化分工产生了一大批具有专业知识的代理人,他们有精力、有能力代理行使好被委托的权利。但在委托代理的关系当中,由于委托人与代理人的效用函数不一样,必然导致两者的利益冲突,在没有有效的制度安排下代理人的行为很可能最终损害委托人的利益。委托代理理论的中心任务就是研究在利益相冲突和信息不对称的环境下,委托人如何设计最优契约激励代理人。 2、案例背景 伯利——米恩斯(1932)在《现代企业与私人财产》中提出了所有权和控制权分离的命题,突破了传统的企业利润最大化的假说,开创从激励角度研究企业之先河。这个命题是委托代理关系的理论背景。 委托代理理论试图模型化如下一类的问题:一个参与人(委托人)想使另一个参与人(代理人)按照前者的利益选择行动,但委托人不能直接观测到代理人选择了什么行动,能观测到的只是另一些变量,这些变量由代理人的行动和其他外生的随机因素共同决定,因而充其量只是代理人行动的不完全信息。委托人的问题是如何根据这些观测到的信息来奖惩代理人,以激励其选择对委托人最有利的行动。 3、模型建立及有关假设条件 3.1模型假设 本文所分析的委托—代理理论模型主要基于以下四个基本假设: (1)在分析委托—代理模型的契约关系时,我们假设存在一个能够确保契约执行的司法体系,并且代理人的行为受到法律的约束。 (2)我们假设委托人和代理人双方同时采用最优的行动以最大化各自的效用函数。

(完整版)委托代理理论综述.

一、委托代理理论概述 (一委托代理关系的产生 1933年,美国学者伯利(Berle和米恩斯(Means在《现代公司与私有财产》一书中,对美国200家大公司进行了分析,发现其中占公司总数量44%、财产的58%的企业是由并未握有公司股权的经理人员控制的,由此他们得出,现代公司的发展,已经发生了“所有与控制”的分离,公司实际上已经为由职业经理组成的“控制者集团”所控制。后来人们把这种现象称为“经理革命”。委托代理关系是随着企业所有权和控制权(经营权的逐步分离而产生的。所谓委托代理关系,就是指委托人把自己的事务交给其代理人代为处理而形成的委托人与代理人之间的责、权、利关系。 (二委托代理理论产生的历史背景 关于资本所有者与企业经营者之间的代理关系,亚当斯密早就有对这一问题的论述,他说:“在钱财的处理上,股份公司的董事为他人尽力,而私人合伙公司的伙员,则纯为自己打算。所以,要想股份公司的董事们监视钱财用途,像私人合伙公司伙员那样用意周到,那是很难做到的。疏忽和浪费,常为股份公司业务经营上多少难免的弊窦。” 委托代理理论是过去30多年里契约理论最重要的发展之一。19世纪60年代末70年代初一些经济学家不满阿罗—德布鲁体系中的企业“黑箱理论”而深入研究企业内部信息不对称和激励问题发展起来的。目前,被大家所公认的委托代理理论的主要创始人包括: 1996年的诺贝尔经济学奖得主英国经济学家莫里斯(Mirrless和2001年的诺贝尔经济学奖三个得主美国经济学家阿克尔洛夫(Akerlof、史宾斯(Spence、斯蒂格利茨(Stiglitz。他们的研究使经济学家们对实际市场经济运行机制的理解有了根本的改进。委托代理理论的中心任务是研究在利益相冲突和信息不对称的环境下,委托人如何设计最优契约激励代理人。 (三委托代理理论的假设前提

销售管理数学建模

销售管理数学建模 销售漏斗是科学反映机会状态以及销售效率的一个重要的销售管理模型。借助销售漏斗,整个销售流程变得更加清晰可控。度爷觉得不错,给网友推荐一下!销售管理的关键节点主要有: ·销售线索:指未经任何验证和过滤的客户联系信息 ·目标识别:销售人员通过各类信息识别,找出对产品或服务有潜在需求的客户·客户意向:确认客户需求,并判断成交意愿度 ·初步认可:对于有意向的客户,销售人员需要开展进一步的交流谈判,看是否能促成订单 ·覆盖关键人物:客户侧的决策组由多人组成,正确识别不同成员的不同角色有利于最后订单的促成 ·商务谈判:对于成交可能性较大的用户,销售人员进行相关的谈判,双方围绕利益点进行价格、产品等洽谈 ·成交:谈判顺利的客户最终成功达成交易 一些小技巧: 【标签】区分客户价值、客户来源等信息

【关联】关联文档中的客户资料及合同信息 【详情】备注客户拜访、沟通记录 如何运用TAPD项目看板搭建销售漏斗【使用案例截图】 一、销售线索: 记录线索名称及接口人 详情:记录客户的基本信息,便于后续跟进。 一般而言客户信息主要包含:公司基本情况、接口人基本信息及联系方式二、目标识别: 从销售线索中识别出值得投入资源跟踪的线索 三、客户意向20%: 经过沟通,筛选出存在需求的客户「描述」记录客户拜访信息及成交意向「标签」区分客户价值 四、初步认可40%: 对于存在需求的客户,销售人员需要开展进一步的洽谈,推动转化为订单五、覆盖关键人物70% 除接口人认可外,还需要赢得决策组其他成员尤其是核心成员的认可

六、商务谈判90%: 围绕利益点开展商业谈判 七、成交合作100%: 度爷在这里祝你成交愉快!销售不一定是打电话,抓住机遇把握时机,成为杰出的市场销售人才。

委托代理理论与业绩评价指标的选择

委托代理理论与业绩评价指标的选择 【摘要】文章从委托代理理论基本模型出发,通过详细讨论模型的假设设定、建立以及结论,进而研究委托代理理论对管理者业绩评价指标选择的启示,并从净利润与股价、财务指标与非财务指标、相对业绩评价指标三个方面对业绩评价指标的选择进行了具体探讨。 【关键词】委托代理理论;不对称信息;薪酬契约;业绩评价指标引言 美国经济学家Berle & Means(1932)在著名的《现代公司与私有财产》一书中开创性地提出“所有权与控制权相分离”的命题,指出:伴随着经济力量的集中,公司股权越来越分散,分散的股权使得所有权与控制权趋于分离。在现代公司中,公司的控制权实际上已经落入管理者手中。 由于所有权与控制权的分离,导致所有者与管理者之间信息不对称,所有者无法观察管理者的努力程度;同时,又由于两者的利益目标不一致,因此产生委托代理问题,特别是隐藏行动的道德风险问题。委托代理理论就是研究如何通过签订薪酬契约来控制道德风险问题,以减少代理成本。通过设计薪酬契约对管理者进行激励,以诱使其按照所有者的目标行事。在薪酬契约中,管理者业绩评价指标的选择至关重要,直接影响着激励机制实施的效果。因此,本文从委托代理理论的基本模型出发,对现代公司制企业中管理者业绩评价指标的选择作一些思考。 从理论的产生和发展过程来看,委托代理理论兴起于20世纪60年代末70年代初(Wilson,1969;Spence & Zeckhavser,1971;Ross,1973;Mirrlees,1974、1975、1976;Holmstrom,1979、1982;Grossman & Hart,1983)。其中,Holmstrom (1979)对委托代理理论的发展作了开创性的工作,为后来的委托代理理论奠定了基础。本文从委托代理理论基本模型的假设设定、模型建立及结论以及对管理者业绩评价的启示三个部分来研究委托代理理论与业绩评价指标的选择。 一、基本委托代理模型的假设设定 (一)符号设定 为叙述方便,本文作下列符号设定: ■表示代理人的行动①(工作努力还是不努力),即代理人在经营中的投入; ■表示自然状态(如市场需求变化等); x表示代理过程中的产出或代理绩效(如利润);

委托代理理论综述

、委托代理理论概述 (一)委托代理关系的产生 1933年,美国学者伯利(Berle )和米恩斯(MeanS在《现代公司与私有 财产》一书中,对美国200家大公司进行了分析,发现其中占公司总数量44% 财产的58%勺企业是由并未握有公司股权的经理人员控制的,由此他们得出,现 代公司的发展,已经发生了“所有与控制”的分离,公司实际上已经为由职业经理组成的“控 制者集团”所控制。后来人们把这种现象称为“经理革命” 代理关系是随着企业所有 。委托权和控制权(经营权)的逐步分离而产生的。所谓委托代理关系,就是指委托人把自 己的事务交给其代理人代为处理而形成的委托人与代理人之间的责、权、利关系。 (二)委托代理理论产生的历史背景 关于资本所有者与企业经营者之间的代理关系,亚当斯密早就有对这一问题的论述,他说:“在钱财的处理上,股份公司的董事为他人尽力,而私人合伙公 司的伙员,贝U纯为自己打算。所以,要想股份公司的董事们监视钱财用途,像私人合伙公司伙员那样用意周到,那是很难做到的。疏忽和浪费,常为股份公司业务经营上多少难免的弊窦。” 委托代理理论是过去30多年里契约理论最重要的发展之一。19世纪60年代末70年代初 一些经济学家不满阿罗一德布鲁体系中的企业“黑箱理论”而深 入研究企业内部信息不对称和激励问题发展起来的。目前,被大家所公认的委托代理 理论的主要创始人包括: 1996年的诺贝尔经济学奖得主英国经济学家莫里斯(Mirrless)和2001年的 诺贝尔经济学奖三个得主美国经济学家阿克尔洛夫(Akerlof)、史宾斯(Spence)、 斯蒂格利茨(Stiglitz)。他们的研究使经济学家们对实际市场经济运行机制的理解有了根本 的改进。委托代理理论的中心任务是研究在利益相冲突和信息不对称的环境下,委托人如何设 计最优契约激励代理人。

委托代理理论

委托代理理论 委托代理理论(Principal-agent Theory) 目录 [隐藏] ? 1 委托代理理论概述 ? 2 委托代理理论基本模型 ? 3 委托代理关系是多次性的动态模型 o 3.1 (一)重复博弈的委托代理模型 o 3.2 (二)代理人市场声誉模型 o 3.3 (三)棘轮效应模型 o 3.4 (四)强制退休的模型 ? 4 委托代理多项任务模型 ? 5 委托人与多个代理人的模型 o 5.1 (一)“打破预算平衡”的模型 o 5.2 (二)考虑逆向选择的模型 o 5.3 (三)合作型模型 ? 6 委托代理理论的相关研究 o 6.1 (一)“相对业绩评估”模型 o 6.2 (二)委托代理关系中,委托人道德风险模型 o 6.3 (三)关于监督问题的模型 o 6.4 (四)最优委托权安排模型 ?7 委托代理理论的理论意义 ?8 相关条目 [编辑] 委托代理理论概述 委托代理理论是建立在非对称信息博弈论的基础上的。非对称信息(asymmetric information)指的是某些参与人拥有但另一些参与人不拥有的信息。信息的非对称性可从以下两个角度进行划分:一是非对称发生的时间,二是非对称信息的内容。从非对称发生的时间看,非对称性可能发生在当事人签约之前(ex ante),也可能发生在签约之后(ex post),分别称为事前非对称和事后非对称。研究事前非对称信息博弈的模型称为逆向选择模型(adverse selection),研究事后非对称信息的模型称为道德风险模型(moral hazard)。从非对称信息的内容看,非对称信息可能是指某些参与人的行为(action),研究此类问题的,我们称为隐藏行为模型(hidden

关于管理绩效_契约_理论和委托代理理论的一种模型_理论部分_

经济科学 1998年第1期 关于管理绩效、“契约”理论和 委托代理理论的一种模型(理论部分) 林长泉 (南京大学国际商学院 210093) 传统的“委托—代理”理论认为:当经理持有的公司股份小于100%时,就会存在着潜在的股东与经理(代理人)之间各自利益的矛盾;与此紧密相关的对管理阶层的绩效考核,也是近年来财务理论关心的首要问题。本文将以一种全新的角度,重新审视这一问题。如果我们的理论和模型能经得住现实数据的考证,那么,在解决此类问题时,我们将会得到一套行之有效的方案。 一、对传统理论的简析 企业性质的新理论,即“契约理论”现已广为流行。它将企业看成一种契约的网络,实际的和内含的合同使企业组织中的所有成员(经营者、所有者、债权人)的作用具体化,并规定了他们各自的权利、义务和在不同的激励或风险条件下的报酬。大多数成员要求有限的风险和稳定的报酬,而企业的所有者则承担所有的剩余风险,正因如此,所有者成为剩余索取者,有权占用各种资产和冲销各种成本后余下的企业盈利。虽然,契约减少了企业内部利益的冲突,但它不可能完全消除由于所有权和控制权(经营权)分离而产生的潜在利益冲突。在现代公司中,所有权一般广为分散,经理阶层事实上对公司具有控制权,所有者们当然希望经理代表其利益行事,然而经理有其自身利益,二者之间往往冲突。为了解决这种由于“委托—代理”而带来的问题,需要增设额外的代理成本:①限制这种经理行为的审计系统。②由经理作出的各种形式的契约保证。③组织系统的变更,以限制经理任意行事的能力。①与此相关,还有两种机制用于控制经理行为。一种是对经理的人力资本作出标价的劳动管理市场,经理市场的竞争导致不称职经理的更换。但是,大多数公司的股权极为分散,从而使得股东们通过董事会来行使他们对公司的控制变得十分困难。股东个人没有很大的投资来支持其花费时间和金钱去密切监控经理行为,这时第二种机制开始作用,即接管公司的市场会起作用。管理不善会导致该公司股价下跌,使公司面临被收购的风险。但是,这是一种“自杀”战术,没有哪位所有者愿意见到公司处于被收购的境地。“经理绩效计划”现在已广泛地被用来作为公司行政管理人员的政策的一部分,此计划有两个目标:①激励经理们竭尽全力充分利用其可控因素以使公司股价最大化。②这些 ①M BA必修核心课程编写组:《理财:资金筹措与使用》,中国国际广播出版社,第5—7页。

销售管理系统UML建模

超市销售系统UML建模

目录引言3 1.1背景3 1.2详细调查4 1.3 编写目的4 1.2预期读者5 1.3产品预期功能5 1.4产品前景5 2 需求分析与用例建模6 2.1可行性分析6 2.1.1管理可行性6 2.1.2经济可行性6 2.1.3技术可行性7 2.1.4社会可行性7 2.2功能需求8 2.3 约束11 2.4系统开发与运行环境11 2.4质量属性11 2.5系统的E-R模型图12 2.6系统功能结构模块图13 2.6系统流程图15 2.6管理业务16 2.6.1组织结构16 2.6.2业务流程调查17 2.6用例建模19 2.6.1确定系统范围和系统边界19 2.6.2确定执行者20 2.6.3确定用例20 2.6.4分层绘制用例图21 3 系统分析与对象类建模27 3.1系统分析原理27 3.2建立对象类28 3.2根据类之间的关系绘制类图31 4 顺序图建模33 5数据流程39 根据调查结果绘出销售系统数据流程图如下:39 6数据字典42 6.1数据流42 6.2逻辑处理43 6.3数据存储43 6.4外部实体44 6.5数据项的表述45 7数据库设计45

总结51 引言 1.1背景 在我国超市形成在20世纪90年代初期,现在已经成为我国零售业的一种重要形态,为国民经济的发展发挥了重要的作用。随着超市高速的发展,其经营管理也变得愈加复杂,早期的售货员站柜台的形式早已不能满足现有销售也的发展,超市需要处理大量的库存信息,还要时刻更新产品的销售信息,不断添加商品信息。面对不同种类的信息,需要合理的数据库结构来保存数据信息,需要有效的程序结构支持各种数据操作的执行。商店自动化的产品管理在欧美等国家早已经实现,也是零售业管理的基础。它最主要的特点是能够实时的和准确的控制店内的销售情况。对于我国来说,超市已经普遍存在,这种自动化管理也成为必要,有必要通过管理系统的应用来提高超市的服务效率。如果可以实时掌握销售流程及销售情况,则可以有效地加速商品的周转率并提高服务质量,而且可以减少产品售价不符等所产生的问题。顾客的消费要求的是希望在超市购物中能基本上都能购得所需的商品,并且还要既保证商品质量还要享受优质,方便的服务。 随着现代科学技术的迅猛发展,计算机技术已经渗透到各个领域,成为各行业必不可少的工具,特别是Internet技术的推广和信息高速公路的建立,使IT 产业在市场竞争中越发显示出其独特的优势,步入信息化时代,有巨大的数据信息等待加工处理和传输,这使得对书数据库的进一步开发和利用显得尤为迫切。作为国内市场的一些中小型超市,它们在信息化过程中的步伐要落后于大中型超市,而对于这些企业的资源管理,信息的存储和处理也显得迫切需要,要适应市场竞争,就需要有高效的处理方式和管理方法,因此加快超市的信息化进程是必可少的。如何将计算机技术与超市管理有机地结合起来,把计算机技术应用于超市管理,更高效地对超市的各种商品信息进行归类、划分、合理地组织,从而减少人们的劳动强度是很有必要的。 本系统针对中小型超市管理特点和技术人员的实际水平,采用Windows易学易用的图形用户界面操作系统,用面向对象的Java语言开发技术和数据库技术,开发出贴近用户实际需求的系统。本系统抛弃了大型财务软件功能泛泛不实用的缺点,既简洁又能满足用户需求,既可以解决在销售过程中出现的问题,也能够用于对商品采购的需求。在减少工作量的基础上,提高超市管理者对企业的

管理学和营销学基础理论和模型

管理学和营销学基础理论和模型 1.营销经典模型:PEST+STP+4P/4C/4R 1.1PEST分析 定义:PEST分析是指宏观环境的分析,P是政治(Political System),E是经济(Economic),S是社会(Social),T是技术(Technological)。在分析一个企业集团所处的背景的时候,通常是通过这四个因素来进行分析企业集团所面临的状况。 图例: 1.2STP分析 定义:STP即目标市场营销,是指企业根据一定的标准对整体市场进行细分后,从中选择一个或者多个细分市场作为自身的目标市场,并针对目标市场进行市场定位。STP分析即市场细分、选择目标市场和产品定位。 图例:

1.34P理论 定义:即产品(product)、价格(price)、促销(promotion)、渠道(place)四要素。(1)产品包含核心产品、实体产品和延伸产品。广义的产品可以是有形的实体,也可以是无形的服务、技术、知识或智慧等。(2)价格的制定手段很多,竞争比较法、成本加成法、目标利润法、市场空隙法,这些方法的目标是使产品成为可交换的商品。(3)传统意义的促销是人员推广、广告、攻关活动和销售促进。这些方式在营销过程中有着非常广泛的应用。(4)渠道是产品从生产方到消费者终端所经历的销售路径。 意义:由密西根大学教授杰罗姆?麦卡锡(E.Jerome Mccarthy)1960年提出,“它的伟大在于它把营销简化并便于记忆和传播”。 图例:

1.44C理论 定义:4C’s的基本原则是以顾客为中心进行企业营销活动规划设计,从产品到如何实现顾客需求(Consumer’s Needs)的满足,从价格到综合权衡顾客购买所愿意支付的成本(Cost),从促销的单向信息传递到实现与顾客的双向交流与沟通(Communication),从通路的产品流动到实现顾客购买的便利性(Convenience)。 意义:从本质上讲,4P’S思考的出发点是企业中心,是企业经营者要生产什么产品、期望获得怎样的利润而制定相应的价格、要将产品怎样的卖点传播和促销、并以怎样的路径选择来销售。这其中忽略了顾客作为购买者的利益特征,忽略了顾客是整个营销服务的真正对象。以客户为中心的新型营销思路的出现,使顾客为导向的4C’S说应运而生。1990年,美国学者劳特朋教授提出了与4P’S相对应的4C’S理论。 图例: 1.54R理论

基于委托代理理论的独立董事模型

Management 经管空间 108 2012年9月 https://www.wendangku.net/doc/9213599294.html, 基于委托代理理论的独立董事模型分析 洛阳市职业技术学院 崔蓓 摘 要:本文运用委托代理理论,重新对独立董事的角色进行了定位,认为独立董事实质上具有代理人的性质,独立董事制度设计应以独立董事信息权保障作为着眼点和出发点。在以上研究的基础上,重新构建独立董事的中国模式,选择代理人成本、能力及其努力程度作为基点,以公司利润作为评价标准来构建独立董事委托代理模型,并依据模型分析提出了独立董事制度重构的基础方案。关键词:委托代理理论 独立董事 模型分析中图分类号:F272 文献标识码:A 文章编号:1005-5800(2012)09(a)-108-02在公司治理结构中,股东(股东大会)是委托人,独立董事接受股东(股东大会)的委托行使企业剩余控制权和决策权是代理人。而高管层是在主要由独立董事构成的董事会的领导下,负责执行董事会的相关决策,负责企业的日常运营,或者只享有部分相对不重要的决策权。委托人(股东)和代理人(独立董事)之间存在的信息不对称,一是委托人如何选择并确定独立董事人选,需要依据代理人(独立董事)传递的信息——职业声誉和职业能力等;二是委托人(股东)如何了解代理人(独立董事)的努力水平,即代理人(独立董事)获取信息的努力水平,以及运用自身才能做出有效决策的能力水平。本文拟在此理论基础上,构建委托代理模型。 1 独立董事委托代理模型分析 在公司的委托代理关系中,考虑风险中性的委托人(股东)和风 险中性的代理人(独立董事),委托人通过设计最优激励方案,委托代理人行使公司管理权。同时,根据激励方案,代理人选择努力水 平。作者认为公司利润的影响因素为代理人努力、 代理人能力类型以及市场竞争度, 那么企业利润为。市场竞争度受产品市场的企业数、产品技术含量等因素的影响。假设是一个连续 变量,其数值越高,企业的利润越低,即 。代理人的努力是私人信息,委托人不易观察到;代理人的能力类型,其中 ,假定服从分布,密度函数为。1.1 风险态度中立的最优激励机制 本文建立独立董事声誉机制,即独立董事职业能力的社会信号传递系统,委托人可以通过代理人的工作经验、职业水平等因素,以及社会评价系统来了解代理人的工作能力,声誉函数。 委托人支付给代理人的工资为, 代理人的保留效用为,代理人的努力成本为,其中,代理人越努力,成本越高, 即 。一般地,努力的边际成本递增,即。代理人能力越高,努力成本与之呈反向变化,即 。同时,努力和能力在效用上是互补的,即,,但是代理人努力程度的边际 利润严格递减,即 ,努力与竞争度在效用上是替代的,即,代理人的能力类型越高,利润越高,即,代理人 的能力类型越高,努力水平越高,声誉越高,即 。根据以上描述,风险中性代理人的效用函数是: (1) 由于委托人(股东)和代理人(独立董事)之间存在的委托代理关系,由于独立董事的独立性,股东对于独立董事的激励目标和动机在于激励独立董事获取充分、有效和真实的信息,以做出有效决策。委托人设计一个激励方案,激励代理人充分利用自身能力获 取并转化信息,即 。可能是c 、和的函数,即。为了便于计算,简写为 ,不影响本文的结论。代理人经过权衡选择适当的能力信息披露,代理人会根据(2)式选择披露水平: (2)写出关于的一阶条件, 在时满足此条件,于是激励相容约束为: 风险中性的委托人的效用函数是: (3)其中代表代理成本, 即为由于代理人和委托人之间信息的不对称,委托人要进行制度设计和执行所要花费的成本。于是,在 激励相容约束和个人理性约束下, 委托人求解模型: (4) 1.2 委托代理模型的解 对于委托人来说,提供一个契约,代理人根据自己的能力水平类型(才能、时间、精力等),选择一个努力水平。 定理1 在不对称信息下,对于 ,如果最优努力的内点解存在,即,则代理人的最优努力满足: (5) 委托人的最优工资满足: 如果努力的内点解不存在,则最优性努力为0,委托人支付给代理人保留效用。 证明:首先,忽略个人理性约束,和是要求的变量, 为了让此问题转化为最优控制问题,首先令,则激励相容约束变为: 本文利用最优化方法求解,先建立H a m i l t o n i a n 函数: (6) 作者简介:崔蓓(1978-),女,河南洛阳人,洛阳市职业技术学院,讲师, 主要从事公司治理、公司财务研究。

相关文档