A. Sathi and M. Fox. Constraint-directed negotiation of resource reallocations. In M. N. Huhns and L. Gasser, editors, Distributed Artificial Intelligence, volume 2 of Research Notes in Artificial Intelligence, chapter 8, pages 163--193. Pitman, 1989.  Home/Search   Document Not in Database   Summary   Related Articles   Check

....priority conventions [10] voting [11] and self modification [5] Previous research provides a foundation for the decomposition and analysis of multi agent conflicts. Researchers have analyzed goal conflicts among fully cooperative [12] and not fully cooperative [13] agents. Resource conflicts, [14] as well as plan conflicts, 15] have also been specifically examined. Domain independent conflict representation and classification is proposed in [16] Conflict detection is an essential process for agents problem solving activities, especially during plan integration. In [17] problem solving ....

A. Sathi and M. S. Fox, "Constraint-directed Negotiation of Resource Reallocations," in Distributed Artificial Intelligence II , L. Gasser and M. N. Huhns, Eds. London: Pitman Publishing, 1989, pp. 163-193.

....However, we use similarity rather than 24 utility to address the inherent uncertainties involved and, as we have shown in section 4.2, this appears to be a better choice in uncertain environments. The process of negotiation has also been modeled as a distributed constraint satisfaction problem [1, 49, 62]. In the work of Sathi and Fox, agents objectives are represented as constraints together with their associated utilities. Strategies (e.g. composition, reconfiguration and relaxation operators) are then used to modify these constraints, or the current solution, until a final solution is reached. ....

A. Sathi and M.S. Fox. Constraint-directed negotiation of resource reallocation. In L. Gasser and M. Huhns, editors, Distributed Artificial Intelligence Volume II, pages 163--195, San Mateo, California, 1989. Morgan Kaufmann. 28

....of the set of missing resources decreases by only one unit after each successful dialogue. 7 Using Dialogue Sequences for Resource Reallocation In this section, we study the use of multiple sequences of dialogues for the solution of the well known and widely studied resource reallocation problem [2, 16]. The problem can be rephrased in our system as follows. Definition 17 (Resource reallocation problem r.r.p. Given an agent system A, with each agent x 2 A equipped with a knowledge base K(x) and an intention I(x) the r.r.p. for an agent x 2 A is the problem of finding a knowledge base ....

....information. We agree that this is very important, especially in contexts such as e commerce and virtual enterprises. The use of negotiation to tackle the resource reallocation problem is not a new idea; among other contributions we cite that by Faratin [3] Sandholm [15] and Sathi and Fox [16]. Most of them study the problem adopting a game theoretic approach and prove results that we consider orthogonal to ours. This paper contributes to a number of issues. The introduction of a logic based framework for negotiation allows us to prove properties that we can use to forecast the ....

A. Sathi and M. Fox. Constraint-directed negotiation of resource reallocation. In L. Gasser and M. Huhns, editors, Distributed Artificial Intelligence, volume II, pages 163--195. Morgan Kaufmann, San Mateo, CA, 1989.

....types and omit t non zero sum games where agents can negotiate over multi issues and both benefit (for example, concession over one issue out of two by each agent) 3.2. 4 CDN Constraint Directed Negotiation (CDN) is an algorithm developed by Sathi and Fox for the problem of resource reallocation [49]. Resource reallocation, or adjustment of initial resources, is performed through buying and selling of resources between agents. The authors have applied CDN to the real world problem of workstation requirements within an engineering organisation. Resources are workstations used by each group ....

A. Sathi and M.S. Fox. Constraint-directed negotiation of resource reallocation. In L. Gasser and M. Huhns, editors, Distributed Artificial Intelligence Volume II, pages 163--195, San Mateo, California, 1989. Morgan Kaufmann.

....However, we use similarity rather than utility to address the inherent uncertainties involved and, as we have shown in section 3.2, this appears to be a better choice in uncertain environments. The process of negotiation has also been modeled as a distributed constraint satisfaction problem [13]. In such cases, an agent s objectives are represented as constraints together with their associated utilities. Strategies (e.g. composition, reconfiguration and relaxation operators) are then used to modify these constraints, or the current solution, until a final solution is reached. The ....

A. Sathi and M. Fox. Constraint-directed negotiation of resource reallocation. In L. Gasser and M. Huhns, editors, Distributed Artificial Intelligence Volume II, pages 163--195, San Mateo, California, 1989. Morgan Kaufmann.

....(1) detects the loan period and overdue notice conflicts, and (2) generates a variety of resolutions. During generation, the analyst chooses resolutions matching those found in the UM analysis. Finally, an integrated specification is derived. 2. A more complete analysis of this example is found in[36]. Figure 1. Oz screen depiction of group requirements and specification development. 4 2.2 Oz Architecture Oz provides support through three basic knowledge based components. Figure 2 illustrates these components. n Planner. The planner takes system requirements as input and outputs a plan ....

Sathi, A., Fox, M., Constraint-directed negotiation of resource reallocations, Eds. L. Gasser, M.N. Huhns, Distributed Artificial Intelligence, Morgan Kaufmann Publishers Inc., San Mateo (1989) 163-193.

....or multiagent negotiation as a distributed search (Durfee Moutyomery 1991; Lander 1994; Lesser 1990) in DAI. In this case, various AI search techniques are proposed together with focused and purposeful communication to improve the efficiency of coordination (Conry et al. 1991; Lander 1994; Sathi Fox 1989). On the other hand, some researchers aim to discover the properties of negotiation under certain formal theoretical frameworks and (quite restricted) assumptions. A large body of work in this camp apply gametheoretical or decision theoretical tools to study how agents should react in a given ....

Sathi, A., and Fox, M. 1989. Constraint-directed negotiation of resource reallocation. In Huhns, M., and Gasser, L., eds., Distributed Artificial Intelligence 2. Pitman. chapter 8, 163--193.

.... systems [Bond Gasser, 1988] Some common conflict resolution strategies include negotiation [Sycara, 1988] arbitration [Steeb et al. 1981] priority conventions [Ioannidis Sellis, 1989] voting [Ephrati Rosenschein, 1991] self modification [Sugie et al. 1995] constraint relaxation [Sathi Fox, 1989], assumption surfacing [Mason Johnson, 1989] evidential reasoning [Carver Lesser, 1995] and standardization and social law [Shoham Tennenholtz, 1992] Given the diversity of available conflict resolution strategies and their various strengths and weaknesses, it has not been demonstrated ....

....conflict [Chu Carroll Carberry, 1995] The action category deals with conflicts among both agents goals and plans. Other researchers have analyzed goal conflicts among cooperative agents [Zlotkin Rosenschein, 1990] and agents that are not fully cooperative [Sycara, 1988] Resource conflicts [Sathi Fox, 1989] as well as plan conflicts [von Martial, 1992] have also been specifically examined. In addition, methods for maintaining coherent agent beliefs have also been investigated [Huhns Bridgeland, 1991] The research presented here leverages these results to provide a detailed classification of ....

[Article contains additional citation context not shown here]

Sathi, A., Fox, M. S. Constraint-directed negotiation of resource reallocations. In L. Gasser & M. N. Huhns (Eds.), Distributed Artificial Intelligence II pp. 163-193. London: Pitman Publishing, 1989.

.... Art In the past two decades, researchers have developed various conflict resolution strategies for multi agent systems [3] Some common conflict resolution strategies include negotiation [28] arbitration [25] priority conventions [13] voting [11] selfmodification [26] constraint relaxation [22], assumption surfacing [19] evidential reasoning [6] and standardization and social law [24] These diverse conflict resolution strategies offer various strengths and weaknesses in different contexts. Similar to Decker s results showing that there is no single best coordination mechanism ....

....analysis of multi agent conflicts. Researchers have The Laboratory for Intelligent Processes and Systems TR98 UT LIPS AGENTS 02 Copyright The University of Texas at Austin 3 analyzed goal conflicts among cooperative agents [31] and agents that are not fully cooperative [27] Resource conflicts [22] as well as plan conflicts [29] have also been specifically examined. In addition, methods for maintaining coherent agent beliefs have also been investigated [12] Chu Carroll and Carberry use two categories (actions and beliefs) to describe, detect, and resolve conflict [7] The action category ....

Sathi, A. and Fox, M. S. 1989, Constraint-directed Negotiation of Resource Reallocations. In Distributed Artificial Intelligence II, Gasser, L. and Huhns, M. N., Eds. (London: Pitman Publishing), 163-193.

....in AI, and there is available work that is of particular relevance from the point of view of CCAF. For instance, there are approaches to reactive and anytime planning (e.g. 5, 6, 12, 13, 15, 22,23, 40] and also [30, 35] real time search (e.g. 25,27, 31] constraint based negotiation (e.g. [41, 44, 47]) negotiation interaction based constraint satisfaction (e.g. 9, 33] anytime constraint satisfaction (e.g. 36, 56, 61] soft contracting that allows agents to break contracts and thus to act more exibly e.g. in response to unexpected changes in their constraints (e.g. 2, 45, 46, 43] ....

A. Sathi and M.S. Fox. Constraint-directed negotiation of resource reallocations. In M.N. Huhns and L. Gasser, editors, Distributed Articial Intelligence, Volume 2, pages 163-193. Pitman/Morgan Kaufmann, Cambridge, MA, 1989.

.... 3 If all possible M contracts including more than three agents and three tasks were to be checked in an interval, that interval would be much longer than the intervals for the other contract types [2] A simpler version of multiagent contracts where bids were grouped into cascades were studied by [14]. 4 The possible contracts depend on the current allocation of This does not mean that a local optimum is necessarily reached in each interval because one contract can enable another contract. If this other contract was tried earlier in the interval, it will not be retried in the same interval. ....

A. Sathi and M. Fox. Constraint-directed negotiation of resource reallocations. In M. N. Huhns and L. Gasser, editors, Distributed Artificial Intelligence, volume 2 of Research Notes in Artificial Intelligence, chapter 8, pages 163--193. Pitman, 1989.

....Whereas the metaphor has been applied successfully in a number of cases, it is sometimes difficult to determine (global) eligibility with only local information. An example of a problem domain which resists a Contract Net approach in this sense is that of resource reallocation. Sathi and Fox [3] claim a mediated approach to be preferable for a resource reallocation problem involving constraints depending on multiple resource offerings. In their approach, a central mediator, supplied with a global view on the problem, is used to overcome the problem of locality of information. Whereas ....

....aware of another agent s individual goals. The problem, resource re allocation, is considerably different, however. This research was supported by the Netherlands Foundation for Scientific Research NWO under grant number 612 322 014. 2 The Problem Context: Resource Reallocation Sathi and Fox [3] situate the resource reallocation problem in the context of a software engineering company comprising teams working on different projects. Each team has a number of resources at its disposal for the duration of a project. Resources, in this context, are computer workstations, whereas the agents ....

[Article contains additional citation context not shown here]

Sathi, A., Fox, M.S.: Constraint-Directed Negotiation of Resource Reallocations. In: L. Gasser and M. Huhns (Eds.), Distributed Artificial Intelligence, Vol. 2, Morgan Kaufmann Publishers (1989)

....portions) to particular agents, such as through market mechanisms. The classical model using this approach is the Contract Net system [2] Similar problems were examined by Durfee and Lesser [3] by Conry [1] and by Ephrati and Rosenschein [4, 6, 5] who used voting techniques) Sathi and Fox [12] used a market like mechanism, where the agents negotiated buying and selling the resources until a compromise is reached. 2 Initial Setting Our initial assumptions closely follow [9] 1. Bilateral Negotiation In each given period of time no more than two agents need the same resource. ....

A. Sathi and M. Fox. Constraint-directed negotiation of resource reallocations. In L. Gasser and M. Huhns, editors, Distributed Artificial Intelligence, volume 2, chapter 8, pages 163--193. Pitman, London, 1989.

....responsible for the scheduling or planning is a natural choice as a mediator. Coordination in the supply chain is necessary when the schedule is found to be infeasible. Scheduling difficulties can often be traced to a particular scarce resource [Smith 89] Fox 90] Sadeh 91] The work of [Sathi 89] indicates that problems with such keystone elements can be solved better by mediated protocols. We now illustrate our approach to coordination with a scenario from the supply chain. 6.3 Supply Chain Management Revisited Recall that, in a manufacturing enterprise, the supply chain consists of ....

Sathi, A. and Fox, M.S. Constraint-Directed Negotiation of Resource Reallocations. Volume 2. Distributed Artificial Intelligence. In Michael N. Huhns and Les Gasser, Pitman Publishing & Morgan Kaufmann Publishers, 1989, pages 163-193, Chapter 8.

....of tasks. 3 In the auction of airwave bandwidth, the Federal Communication Commission used a simultaneous ascending auction to provide for the bidders the possibility to cluster the frequencies for which they were bidding, without explicit cluster contracts (McAfee McMillan 1996) 4 Sathi and Fox (1989) (Sathi Fox 1989) studied a simpler version of multiagent contracts where bids were grouped into cascades. timum is reached in a finite number of steps (contracts) when used among myopically individually rational agents, irrespective of the order in which contracts are proposed and made ....

.... In the auction of airwave bandwidth, the Federal Communication Commission used a simultaneous ascending auction to provide for the bidders the possibility to cluster the frequencies for which they were bidding, without explicit cluster contracts (McAfee McMillan 1996) 4 Sathi and Fox (1989) (Sathi Fox 1989) studied a simpler version of multiagent contracts where bids were grouped into cascades. timum is reached in a finite number of steps (contracts) when used among myopically individually rational agents, irrespective of the order in which contracts are proposed and made (Sandholm ....

Sathi, A., and Fox, M. 1989. Constraint-directed negotiation of resource reallocations. In Huhns, M. N., and Gasser, L., eds., Distributed Artificial Intelligence, volume 2 of Research Notes in Artificial Intelligence.

....evaluation, decision process theory, and AI planning. This theoretical foundation allows us to incrementally acquire and change preferences during 2. This can be contrasted with Distributed AI (DAI) research which concentrates on architecture while simplifying resolution generation [5][18][20] 17] 16] 4 negotiations. In contrast, other approaches are based on knowing all preferences and resolutions before the start of negotiation[17] 16] These approaches do not automate the generation (or conflict detection) process. Doing so would require overcoming the preference bottleneck, ....

Sathi, A., Fox, M., Constraint-directed negotiation of resource reallocations, Eds. L. Gasser, M.N. Huhns, Distributed Artificial Intelligence, Morgan Kaufmann Publishers Inc., San Mateo (1989) 163-193.

....where walrasian mechanisms fail. 1 TOOL REALLOCATION We investigate auction based mechanisms for a class of problems that can be described as tool reallocation problems . A tool is a resource that is essentially indivisible. Typical examples of indivisible resource types are computer workstations[5], oil tankers, and low capacity communication channels. The essential characteristic here is that agents generally require only a small number of resources of the types they desire. Tool reallocation problems constitute a subclass of resource reallocation problems, for which mechanism design is ....

....are communicated via a central auctioneer agent, rather than directly between agents. As we shall show later, this proves to be a fortunate decision with respect to global performance. As far as the problem is concerned, our research is inspired by the workstation reallocation problem sketched in [5]. Their Mediated Negotiation mechanism, however, differs from our Informed Gambling approach, in that it features a relatively sophisticated central mediator. 4 THE AUCTION METAPHOR The auction metaphor is used explicitly in literature on the allocation of processor and memory access to ....

A. Sathi and M. S. Fox, `Constraint-directed negotiation of resource reallocations', in Distributed Artificial Intelligence, eds., L. Gasser and M. Huhns, volume 2, Pitman, London, (1989).

.... feedback about what they like or don t like about particular proposals, resulting in a progressively better understanding of the shared requirements for solutions over time 1 Sathi similarly uses the term composition as the name of a specific search operator that combines local information [Sathi and Fox, 1989] ffl agents can play different roles in the negotiation process, e.g. an agent can be a reviewer for another agent s proposal and then be a generator for a counter proposal ffl conflicts exist among the agents requirements for acceptable solutions ffl agents incrementally relax their ....

.... to conflict resolution and avoidance [ Adler et al. 1989, Klein, 1991, Lander and Lesser, 1992a, Sycara, 1985, Werkman, 1992 ] task allocation [ Cammarata et al. 1983, Durfee and Montgomery, 1990, Davis and Smith, 1983 ] and resource allocation [ Adler et al. 1989, Conry et al. 1992, Sathi and Fox, 1989, Sycara et al. 1991 ] Negotiation is sometimes treated as an independent process that is used to select one of a set of existing alternative solutions [ Zlotkin and Rosenschein, 1990 ] rather than as an inherent part of a solution generation process. It can be difficult under conditions where ....

[Article contains additional citation context not shown here]

Arvind Sathi and Mark S. Fox. Constraint-directed negotiation of resource reallocations. In Les Gasser and Michael Huhns, editors, Distributed Artificial Intelligence, Volume 2, pages 163--193. Pitman, Morgan Kaufmann Publishers, 1989.

....of tasks. 3 In the auction of airwave bandwidth, the Federal Communication Commission used a simultaneous ascending auction to provide for the bidders the possibility to cluster the frequencies for which they were bidding, without explicit cluster contracts (McAfee McMillan 1996) 4 Sathi and Fox (1989) (Sathi Fox 1989) studied a simpler version of multiagent contracts where bids were grouped into cascades. timum is reached in a finite number of steps (contracts) when used among myopically individually rational agents, irrespective of the order in which contracts are proposed and made ....

.... In the auction of airwave bandwidth, the Federal Communication Commission used a simultaneous ascending auction to provide for the bidders the possibility to cluster the frequencies for which they were bidding, without explicit cluster contracts (McAfee McMillan 1996) 4 Sathi and Fox (1989) (Sathi Fox 1989) studied a simpler version of multiagent contracts where bids were grouped into cascades. timum is reached in a finite number of steps (contracts) when used among myopically individually rational agents, irrespective of the order in which contracts are proposed and made (Sandholm ....

Sathi, A., and Fox, M. 1989. Constraint-directed negotiation of resource reallocations. In Huhns, M. N., and Gasser, L., eds., Distributed Artificial Intelligence, volume 2 of Research Notes in Artificial Intelligence.

....Also, identifying profitable OCSM contracts can be difficult in the large. Therefore, in large scale problem instances it is important to be able trade off solution quality against negotiation time. For example, the agents may want to find the best solution that is obtainable in a given amount 3 Sathi and Fox (1989) (Sathi Fox 1989) studied a simpler version of multiagent contracts where bids were grouped into cascades. of time. This paper studies how the different contract types affect that tradeoff. Contracting system In principle our contracting system implementation can be used to solve ....

....OCSM contracts can be difficult in the large. Therefore, in large scale problem instances it is important to be able trade off solution quality against negotiation time. For example, the agents may want to find the best solution that is obtainable in a given amount 3 Sathi and Fox (1989) (Sathi Fox 1989) studied a simpler version of multiagent contracts where bids were grouped into cascades. of time. This paper studies how the different contract types affect that tradeoff. Contracting system In principle our contracting system implementation can be used to solve reallocation problems with any ....

Sathi, A., and Fox, M. 1989. Constraint-directed negotiation of resource reallocations. In Huhns, M. N., and Gasser, L., eds., Distributed Artificial Intelligence, volume 2, Pitman, chapter 8, 163--193.

....that is, all possible contracts have been tried but none have been performed. In the following subsections the order of trying different contracts within each contract type is discussed. The agents are enumerated from 1 to jAj, and each agent s tasks from 1 to jT i j. 6 Sathi and Fox (1989) [19] studied a simpler version of multiagent contracts where bids were grouped into cascades. Sequencing of Original Contracts An O contract allows one agent to move one task to one other agent. In our experiments we sequenced the O contracts as follows. First, agent 1 s tasks are attempted to be ....

A. Sathi and M. Fox. Constraint-directed negotiation of resource reallocations. In M. N. Huhns and L. Gasser, editors, Distributed Artificial Intelligence, volume 2 of Research Notes in Artificial Intelligence, chapter 8, pages 163--193. Pitman, 1989.

....multi agent, temporal, dynamically capable ones. A community of researchers working on distributed artificial intelligence (DAI) has arisen (for a survey of DAI see [1, 15] One of the most difficult subjects that has occupied the efforts of the DAI community has been the subject of negotiation [55, 19, 59, 41, 5, 9, 47, 51, 34, 64, 33, 11, 35, 61]. Davis and Smith s work on the Contract Net [55] introduced a form of simple negotiation among cooperative agents, with one agent announcing the availability of tasks and awarding them to other bidding agents. Malone refined this technique considerably by overlaying it with a more sophisticated ....

.... general task, or in which the agents belong to the same organization or unit (see for example [19] which describes a method for synthesizing multi agent plans from simple single agent plans, 52] which deals with project management, 9] and [10] which deal with the vehicle monitoring domain, and [51] which deals with resource reallocations) Conflicts among the agents in these environments may arise while each tries to achieve its own sub tasks (for example, they may need to share the same resources) but their overall task is the same. Our work takes as a point of departure the work of ....

A. Sathi and M.S. Fox. Constraint-directed negotiation of resource reallocations. In L. Gasser and M. N. Huhns, editors, Distributed Artificial Intelligence, Volume II, pages 163--194. Pitman/Morgan Kaufmann, London, 1989.

....and agreeing that the contract becomes valid only if every agent signs. 2. Multiagent contracts through mutual contracts. A multiagent contract is equivalent to a sequence of mutual contracts. In cases where a local optimum with respect to mutual contracts has been reached, 11 Sathi et al. [Sathi and Fox, 1989] did this by having a centralized mediator cluster several announcements and bids from multiple agents into atomic contracts. That is unreasonable if decentralization is desired. the first mutual contracts in the sequence will incur losses. Thus, one or more agents have to incur risk in ....

A Sathi and M Fox. Constraintdirected negotiation of resource reallocations. In Michael N. Huhns and Les Gasser, eds., Distributed Artificial Intelligence, vol. 2 of Research Notes in Artificial Intelligence, ch. 8, pages 163--193. Pitman, 1989.

....capabilities be applied to select which constraints to relax when conflicts do occur. Although we do not address specific techniques for conflict management in this article, it is an important and encompassing problem. Work describing computational conflict management techniques includes [1, 12, 15, 21, 22, 26, 27]. 3.1 Information Shared in the STEAM System The information shared in these experiments was limited to simple boundary constraints of the form fconstraint form, flexibilityg. A constraint form is a single clause with the syntax (x n) x = n) x n) or (x = n) as ....

A. Sathi and M.S. Fox. Constraint-directed negotiation of resource reallocations. In Les Gasser and Michael Huhns, editors, Distributed Artificial Intelligence, Volume 2, chapter 8, pages 163--193. Pitman Publishing, London, 1989.

....to execute at each agent. A negotiated search strategy is designed to reduce the amount of search required to find mutually acceptable solutions by taking advan 1 Sathi similarly uses the term composition as the name of a specific negotiated search operator that combines local information [18]. tage of specific attributes of both individual agents and agent sets. The use of a particular strategy requires that certain capabilities, characteristics, and relationships exist among some agents in the agent set. This is a problem that is unique to heterogeneous multi agent systems where ....

....In heterogeneous agent problem solving, each agent can analyze its local solution space but there is no effective method for analyzing the shared space in a general way since no agent has enough knowledge about the constraints of other agents. This problem has been addressed somewhat by Sathi [18], Sycara [21] and Mammen [15] in their work on constraint directed negotiation. However, these investigations of constraint directed reasoning in multi agent systems have required that the agents share an underlying integrated problem solving methodology and agent architecture. Although these ....

Arvind Sathi and Mark S. Fox. Constraint-directed negotiation of resource reallocations. In Les Gasser and Michael Huhns, editors, Distributed Artificial Intelligence, Volume 2, pages 163--193. Pitman, Morgan Kaufmann Publishers, 1989.

....develop reveal opportunities to negotiate semi cooperative deals even in conflicting situations. The notion of semi cooperative deals subsumes other types of deals, and forms the core of the authors proposed Unified Negotiation Protocol that provides a formal framework for studying negotiation [18, 51, 80, 88]. DAI research on negotiation and coordination has generally emphasized the need for each agent in a multiagent environment to modify what it does to improve group performance. This view typically assumes that the set of interacting agents is fixed, and so better coordination requires the agents ....

Arvind Sathi and Mark S. Fox. Constraint-directed negotiation of resource reallocations. In Michael N. Huhns and Les Gasser, editors, Distributed Artificial Intelligence, volume 2 of Research Notes in Artificial Intelligence. Pitman, 1989.

....have reached a local optimumwith respect to mutual (O , C , and S ) contracts of any size, solution enhancements may be possible if tasks were transferred among more than two agents. 8 8 Sathi and Fox studied the role of grouping buy and sell bids into cascades which may involve multiple agents (Sathi Fox 1989). Their setting is simpler than ours in that the value of a contract to an agent does not depend on which other ones of the agent s bids get accepted. In our setting, an agent s valuation of a contract depends significantly on which other ones of the agent s bids get accepted (marginal cost ....

Sathi, A., and Fox, M. 1989. Constraint-directed negotiation of resource reallocations. In Huhns, M. N., and Gasser, L., eds., Distributed Artificial Intelligence, volume 2 of Research Notes in Artificial Intelligence. Pitman.

....of a cooperative dialogue among agents to resolve the uncertainty and incompleteness of an agent s knowledge base. These are highly directed dialogues that attempt to minimize the communication necessary to arrive at consistent solutions to interdependent subproblems and implement network control. [1, 3, 6, 10, 21, 22, 27, 28, 29] 3. Tolerance of inconsistency: An agent s problem solving is structured to operate with local knowledge bases that can be incomplete, inconsistent, and out of date. Error resolution is an integral part of network problem solving and agents do the best they can with their current information. 9, ....

Arvind Sathi and Mark S. Fox. Constraint-directed negotiation of resource reallocations. In Distributed Artificial Intelligence, volume 2, pages 163--193. Morgan Kaufmann, California, 1989.

....soon become intractable for large games. The ADEPT negotiation model has also been influenced by social science models of negotiation. These models do not make the restrictive assumptions of game theory and they attempt to identify and describe behaviours that may achieve satisfactory outcomes [6, 22, 38, 42] (rather than prescribe behaviour like the game theory models) For example, the competitive, accommodative and conciliatory negotiation behaviours in ADEPT are heuristics that agents use as search operators to prune their set of possible actions. However, although such models are inspired by ....

A. Sathi and M. S. Fox, Constraint directed negotiation of resource re-allocations. In Distributed Artificial Intelligence II (eds. L. Gasser and M. Huhns), Pitman, 163-193, 1989.

..... Factory Agents: each factory has an agent that is the factory level scheduler. 3.0 A Mediated Approach Much coordination work looks at negotiation among agents as the coordination mechanism [Lesser 81] Durfee 87a] Durfee 87b] Durfee 91] In contrast we adopt a mediated approach. Sathi 89] shows that in some cases a mediated solution can be significantly better than a negotiated solution. The problem investigated is resource reallocation where each agent has some resources and needs other resources. Agents sell the resources they have for those they need. Experimental results show ....

....can often be traced to scarce or bottleneck resources [Smith 89] Sadeh 91] Fox 90] Coordination in the supply chain is needed when the schedule is found to be infeasible due to environmental events, therefore we expect difficulties will arise from similar resource properties. The work of [Sathi 89] indicates that problems with such keystone elements can be solved much better by mediated protocols. 4. Texture measurements [Fox 89] Sycara 91] assess structural properties of the constraint graph representation of the problem. Based on these measurements, heuristic search decisions are ....

Sathi, A. and Fox, M.S. Constraint-Directed Negotiation of Resource Reallocations. In Michael N. Huhns and Les Gasser (Eds.), Distributed Artificial Intelligence. Volume 2. Pitman Publishing & Morgan Kaufmann Publishers, 1989, pages 163-193, Chapter 8.

....found to be infeasible. Given that a schedule forms the core of this graph, we conjecture that conflicts arise for reasons similar to those that lead to scheduling difficulties. Often scheduling difficulties can be traced to a particular scarce resources [Ow 87] Fox 90] Sadeh 91] The work of [Sathi 89] indicates that problems with such keystone elements can be solved better by mediated protocols. The graph representing the commitments between agents is built by the mediator starting with the current schedule. The mediator requests information from the other agents which is adds to the graph to ....

Sathi, A. and Fox, M.S. Constraint-Directed Negotiation of Resource Reallocations. Volume 2. Distributed Artificial Intelligence. In Michael N. Huhns and Les Gasser, Pitman Publishing & Morgan Kaufmann Publishers, 1989, pages 163-193, Chapter 8.

No context found.

A. Sathi and M. Fox. Constraint-directed negotiation of resource reallocations. In M. N. Huhns and L. Gasser, editors, Distributed Artificial Intelligence, volume 2 of Research Notes in Artificial Intelligence, chapter 8, pages 163--193. Pitman, 1989.

No context found.

Arvind Sathi and Mark S. Fox. Constraint-directed negotiation of resource reallocations. In Michael N. Huhns and Les Gasser, editors, Distributed Artificial Intelligence, volume 2 of Research Notes in Artificial Intelligence. Pitman, 1989.

No context found.

Arvind Sathi and Mark S. Fox. Constraint-directed negotiation of resource reallocations. In Michael N. Huhns and Les Gasser, editors, Distributed Artificial Intelligence, volume 2 of Research Notes in Artificial Intelligence. Pitman, 1989.

No context found.

A Sathi and Mark Fox. Constraintdirected negotiation of resource reallocations. In Michael N. Huhns and Les Gasser, editors, Distributed Artificial Intelligence, volume 2 of Research Notes in Artificial Intelligence, chapter 8, pages 163--193. Pitman, 1989.

No context found.

A. Sathi and M.S. Fox. Constraint-directed negotiation of resource reallocations. In M.N. Huhns and L. Gasser, editors, Distributed Artificial Intelligence, Volume 2, pages 163--193. Pitman/Morgan Kaufmann, Cambridge, MA, 1989.

No context found.

Sathi, A., Fox, M., Constraint-directed negotiation of resource reallocations, Eds. L. Gasser, M.N. Huhns, Distributed Artificial Intelligence, Morgan Kaufmann Publishers Inc., San Mateo (1989) 163-193.

No context found.

A. Sathi and M.S. Fox. Constraint-Directed Negotiation of Resource Reallocations. In [Gasser and Huhns 1989, p.163--194].

Online articles have much greater impact   More about CiteSeer.IST   Add search form to your site   Submit documents   Feedback  

CiteSeer.IST - Copyright Penn State and NEC