System and method for a dynamic auction with package bidding

• US 20040054551A1
• Filed: 05/22/2003
• Published: 03/18/2004
• Est. Priority Date: 11/22/2000
• Status: Active Grant

• Current Assignee: Efficient Auctions, LLC
• Original Assignee: Efficient Auctions, LLC
• Inventors: Milgrom, Paul R., Ausubel, Lawrence M
• Granted Patent: US 8,566,211 B2
• Time in Patent Office: Days
• US Class Current: 705/1
• CPC Class Codes: G06Q 30/0601 Electronic shopping [e-shopping]
• CPC Class Codes: G06Q 30/08 Auctions
• CPC Class Codes: G06Q 40/04 Trading; Exchange, e.g. stocks, commodities, derivatives or currency exchange

Abstract

The present invention is a system and method for a computer-implemented auction, based on a system with multiple bidder and auctioneer computers or terminals (20a-n,30) communicating with the server (10) over a network (40). The computers (20a-n) are employed by bidders, the computer (30) is employed by the auctioner, and the server (10) is the acution computer. That server consist of a CPU (11), memory (12), a data storage device (13), a communication interface (140), a clock (150), an operating system (16), and an acution program (17). In one embodiment, the system architectura is as a client-server system: the acution computer is a server; and the bidder and auctioneer are clients.

Claims

1. A computer system for conducting an auction of a set of items among a plurality of bidders wherein bids are received and an allocation of the items to the bidders is determined by the computer system based on the bids, comprising:

means for receiving bids, and

processing means for processing the bids to determine an allocation of the items among the bidders, wherein the processing means includes;

means for determining whether at least one combination of new bids from a given bidder satisfies a relation with respect to prior bids of the given bidder,

entering means for entering all of the new bids from the given bidder if at least one combination of new bids from the given bidder satisfies said relation, and entering no new bids from the given bidder if no combination of new bids from the given bidder satisfies said relation.

2. A computer system as recited in claim 1 wherein the means for determining responds to improvement in bid quality of the bidder'"'"'s new bids compared to the bid quality of the bidder'"'"'s prior bids.

3. A computer system as recited in claim 2 wherein the means for determining requires the bid quality of the bidder'"'"'s new bids to exceed the quality of the bidder'"'"'s prior bids by a predetermined quantity.

4. A computer system as recited in claim 3 wherein the predetermined quantity is fixed.

5. A computer system as recited in claim 3 wherein the predetermined quantity is related to the state of the auction system.

6. A computer system as recited in claim 1 wherein the means for determining requires the bid quality of the bidder'"'"'s new bids to exceed the quality of the bidder'"'"'s prior bids by a predetermined quantity unless the bidder is a provisional winner.

7. A computer system as recited in claim 6 wherein the predetermined quantity is fixed.

8. A computer system as recited in claim 6 wherein the predetermined quantity is related to the state of the auction system.

9. A computer system as recited in claim 1 wherein the bids are package bids.

10. A computer system as recited in claim 9 wherein single package bids represent the combinations of new bids.

11. A computer system as recited in claim 1 wherein pluralities of bids represent the combinations of new bids.

12. A computer system as recited in claim 9 wherein the means for determining responds to improvement in bid quality of the bidder'"'"'s new bids compared to the bid quality of the bidder'"'"'s prior bids.

13. A computer system as recited in claim 2 wherein the means for determining includes means to detect bid quality improvement including:

means for determining a relative value index for each package in the auction from a state of the auction,

means for determining the quality of each bid by the bidder as the excess of the bid over the relative value index for the package, and

means for determining the bid quality index for the bidder as a function of the excess of each bid over the relative value index for the package.

14. A computer system as recited in claim 13 wherein the function of the excess of each bid over the relative value index for the package is the maximum excess of a bid over the relative value index for the package.

15. A computer system as recited in claim 2 wherein the means for determining includes means to detect bid quality improvement including:

means for determining a relative value index for each package in the auction from a state of the auction,

means for determining the quality of each combination of bids by the bidder as the excess of the combination of bids over the relative value index for the associated package, and

means for determining the bid quality index for the bidder as a function of the excess of each combination of bids over the relative value index for the associated package.

16. A computer system as recited in claim 15 wherein the function of the excess of each combination of bids over the relative value index for the associated package is the maximum excess of a combination of bids over the relative value index for the associated package.

17. A computer system as recited in claim 2 including means for accepting or rejecting new bids by the bidder in response to the entering means.

18. A computer system as recited in claim 2 which further includes an agent for each bidder, where each of the bids are proxy bids generated by the agent and wherein the agent is responsive to the means for determining to generate a new proxy bid or refuse to generate a new proxy bid.

19. A computer implemented method for conducting an auction of a set of items among a plurality of bidders wherein bids are received at a computer and an allocation of the items to the bidders is determined by the computer based on the bids, comprising:

receiving bids, and

processing the bids to determine an allocation of the items among the bidders, wherein processing the bids includes;

determining whether at least one combination of new bids from a given bidder satisfies a relation with respect to prior bids of the given bidder, entering all of the new bids from the given bidder if at least one combination of new bids from the given bidder satisfies said relation, and entering no new bids from the given bidder if no combination of new bids from the given bidder satisfies said relation.

20. A computer implemented method as recited in claim 19 wherein the determination is based on improvement in bid quality of the bidder'"'"'s new bids compared to the bid quality of the bidder'"'"'s prior bids.

21. A computer implemented method as recited in claim 20 wherein the determination requires the bid quality of the bidder'"'"'s new bids to exceed the quality of the bidder'"'"'s prior bids by a predetermined quantity.

22. A computer implemented method as recited in claim 21 wherein the predetermined quantity is fixed.

23. A computer implemented method as recited in claim 21 wherein the predetermined quantity is related to auction state information.

24. A computer implemented method as recited in claim 19 wherein the determination requires the bid quality of the bidder'"'"'s new bids to exceed the quality of the bidder'"'"'s prior bids by a predetermined quantity unless the bidder is a provisional winner.

25. A computer implemented method as recited in claim 24 wherein the predetermined quantity is fixed.

26. A computer implemented method as recited in claim 25 wherein the predetermined quantity is related to auction state information. [In system claim, make this “

a state of the auction system”

]

27. A computer implemented method as recited in claim 19 wherein the bids are package bids.

28. A computer implemented method as recited in claim 27 wherein single package bids represent the combinations of new bids.

29. A computer implemented method as recited in claim 19 wherein pluralities of bids represent the combinations of new bids.

30. A computer implemented method as recited in claim 27 wherein the determination is based on improvement in bid quality of the bidder'"'"'s new bids compared to the bid quality of the bidder'"'"'s prior bids.

31. A computer implemented method as recited in claim 20 wherein bid quality improvement is determined by

determining from a state of the auction, a relative value index for each package in the auction,

based on the relative value index for each package, determining the quality of each bid by the bidder as the excess of the bid over the relative value index for the package, and determining the bid quality index for the bidder as a function of the excess of each bid over the relative value index for the package.

32. A computer implemented method as recited in claim 31 wherein the function of the excess of each bid over the relative value index for the package is the maximum excess of a bid over the relative value index for the package.

33. A computer implemented method as recited in claim 20 wherein bid quality improvement is determined by

determining from a state of the auction, a relative value index for each package in the auction,

based on the relative value index for each package, determining the quality of each combination of bids by the bidder as the excess of the combination of bids over the relative value index for the associated package, and determining the bid quality index for the bidder as a function of the excess of each combination of bids over the relative value index for the associated package.

34. A computer implemented method as recited in claim 33 wherein the function of the excess of each combination of bids over the relative value index for the associated package is the maximum excess of a combination of bids over the relative value index for the associated package.

35. A computer implemented method as recited in claim 20 wherein the computer enforces the determination by accepting or rejecting new bids by the bidder.

36. A computer implemented method as recited in claim 20 where each bidder operates an agent and each of the bids are proxy bids generated by the agent and wherein the determination is enforced by an agent generating new proxy bids or refusing to generate a new proxy bid.

37. A computer system for conducting an auction of a set of items among a plurality of bidders wherein bids are received at a computer and an allocation of the items to the bidders is determined by the computer based on the bids, comprising:

means for receiving bids, and

means for processing the bids to determine an allocation of the items among the bidders, which includes;

means for determining whether or not to accept new bids from a bidder based on a relation between the prior bids by the bidder and the new bids.

38. A computer system as recited in claim 37 wherein the means for determining includes means responding to improvement in bid quality of the bidder'"'"'s new bids compared to the bid quality of the bidder'"'"'s prior bids.

39. A computer system as recited in claim 38 wherein the means for determining includes means to respond to the bid quality of the bidder'"'"'s new bids exceeding the quality of the bidder'"'"'s prior bids by a predetermined quantity.

40. A computer system as recited in claim 39 wherein the predetermined quantity is fixed.

41. A computer system as recited in claim 39 wherein the predetermined quantity is related to a state of the auction.

42. A computer system as recited in claim 37 wherein the means for determining responds to the bid quality of the bidder'"'"'s new bids exceeding the quality of the bidder'"'"'s prior bids by predetermined quantity unless the bidder is a provisional winner.

43. A computer system as recited in claim 42 wherein the predetermined quantity is fixed.

44. A computer system as recited in claim 42 wherein the predetermined quantity is related to the state of the auction.

45. A computer system as recited in claim 37 wherein the bids are package bids.

46. A computer system as recited in claim 45 wherein the means for determining responds to improvement in bid quality of the bidder'"'"'s new bids compared to the bid quality of the bidder'"'"'s prior bids.

47. A computer system as recited in claim 38 wherein the means for determining includes:

relative value computing means for computing from a state of the auction, a relative value index for each package in the auction,

bid quality means for determining the quality of each bid by the bidder as the excess of the bid over the relative value index for the package, and

bid quality index means for determining the bid quality index for the bidder as a function of the excess of each bid over the relative value index for the package.

48. A computer system as recited in claim 47 wherein the bid quality index means includes means for determining the function of the excess of each bid over the relative value index for the package as the maximum excess of a bid over the relative value index for the package.

49. A computer system as recited in claim 38 wherein the means for determining includes:

relative value computing means for computing from a state of the auction, a relative value index for each package in the auction,

bid quality means for determining the quality of each combination of bids by the bidder as the excess of the combination of bids over the relative value index for the associated package, and

bid quality index means for determining the bid quality index for the bidder as a function of the excess of each combination of bids over the relative value index for the associated package.

50. A computer system as recited in claim 49 wherein the bid quality index means includes means for determining the function of the excess of each combination of bids over the relative value index for the associated package as the maximum excess of a combination of bids over the relative value index for the associated package.

51. A computer system as recited in claim 38 which further includes at least one agent and each of the bids are proxy bids generated by the agent.

52. A computer implemented method for conducting an auction of a set of items among a plurality of bidders wherein bids are received at a computer and an allocation of the items to the bidders is determined by the computer based on the bids, comprising:

receiving bids, and

processing the bids to determine an allocation of the items among the bidders, wherein processing the bids includes;

determining whether or not to accept new bids from a bidder based on a relation between the prior bids by the bidder and the new bids.

53. A computer implemented method as recited in claim 52 wherein the determination is based on improvement in bid quality of the bidder'"'"'s new bids compared to the bid quality of the bidder'"'"'s prior bids.

54. A computer implemented method as recited in claim 53 wherein the determination requires the bid quality of the bidder'"'"'s new bids to exceed the quality of the bidder'"'"'s prior bids by a predetermined quantity.

55. A computer implemented method as recited in claim 54 wherein the predetermined quantity is fixed.

56. A computer implemented method as recited in claim 54 wherein the predetermined quantity is related to a state of the auction.

57. A computer implemented method as recited in claim 52 wherein the determination requires the bid quality of the bidder'"'"'s new bids to exceed the quality of the bidder'"'"'s prior bids by predetermined quantity unless the bidder is a provisional winner.

58. A computer implemented method as recited in claim 57 wherein the predetermined quantity is fixed.

59. A computer implemented method as recited in claim 57 wherein the predetermined quantity is related to the state of the auction.

60. A computer implemented method as recited in claim 52 wherein the bids are package bids.

61. A computer implemented method as recited in claim 60 wherein the determination is based on improvement in bid quality of the bidder'"'"'s new bids compared to the bid quality of the bidder'"'"'s prior bids.

62. A computer implemented method as recited in claim 53 wherein bid quality improvement is determined by

determining from a state of the auction, a relative value index for each package in the auction,

based on the relative value index for each package, determining the quality of each bid by the bidder as the excess of the bid over the relative value index for the package, and determining the bid quality index for the bidder as a function of the excess of each bid over the relative value index for the package.

63. A computer implemented method as recited in claim 62 wherein the function of the excess of each bid over the relative value index for the package is the maximum excess of a bid over the relative value index for the package.

64. A computer implemented method as recited in claim 53 wherein bid quality improvement is determined by

determining from a state of the auction, a relative value index for each package in the auction,

based on the relative value index for each package, determining the quality of each combination of bids by the bidder as the excess of the combination of bids over the relative value index for the associated package, and determining the bid quality index for the bidder as a function of the excess of each combination of bids over the relative value index for the associated package.

65. A computer implemented method as recited in claim 64 wherein the function of the excess of each combination of bids over the relative value index for the associated package is the maximum excess of a combination of bids over the relative value index for the associated package.

66. A computer implemented method as recited in claim 53 wherein the computer enforces the determination by accepting or rejecting new bids by the bidder.

67. A computer implemented method as recited in claim 53 wherein each bidder operates an agent and each of the bids are proxy bids generated by the agent and wherein the determination is enforced by an agent generating new proxy bids or refusing to generate a new proxy bid.

68. A computer system for conducting an auction for a set of items among a plurality of bidders wherein bids are received at the computer system and items are allocated to the bidders by the computer system, comprising

means for receiving bids, and

processing means for processing bids to allocate the items among the bidders, wherein the processing means includes:

means for selecting, for a bidder, a subset of all of the bidder'"'"'s bids at a given time, means for ranking the selected bids in decreasing order of quality,

means for retaining the N highest-ranked selected bids, where N is an integer greater than unity, and for discarding the selected bids other than the N highest-ranked selected bids.

69. A computer system as recited in claim 68 wherein the bids are package bids.

70. A computer system as recited in claim 69 wherein the set of all of the bidder'"'"'s bids at the given time includes bids previously subject to selection and ranking as well as bids not previously subject to selection and ranking.

71. A computer system as recited in claim 70 wherein the means for ranking, ranks the quality of selected bids based on excess value of the bids over a relative value index for the associated package.

72. A computer system as recited in claim 71 wherein the relative value index is a measure of value attached to each package of the set of all items, based on the bids at the given time.

73. A computer system as recited in claim 68 wherein the selecting means selects a subset of bids which comprise all of the bids by the bidder which are in effect.

74. A computer system as recited in claim 68 wherein the selecting means selects a subset of bids which comprise all of the bids by the bidder which are in effect except for provisionally-winning bids.

75. A computer implemented method for conducting an auction for a set of items among a plurality of bidders wherein bids are received at a computer and items are allocated to the bidders by the computer, comprising

receiving bids, and

processing bids to allocate the items among the bidders, wherein processing the bids includes, at a given time:

selecting, for a bidder, a subset of all of the bidder'"'"'s bids,

ranking the selected bids in decreasing order of quality,

retaining the N highest-ranked selected bids, where N is an integer greater than unity, and discarding the selected bids other than the N highest-ranked selected bids.

76. A computer implemented method as recited in claim 75 wherein the bids are package bids.

77. A computer implemented method as recited in claim 76 wherein the set of all of the bidder'"'"'s bids at the given time includes bids previously subject to selection and ranking as well as bids not previously subject to selection and ranking.

78. A computer implemented method as recited in claim 77 wherein the quality of selected bids is based on excess value of the bids over a relative value index for the associated package.

79. A computer implemented method as recited in claim 78 wherein the relative value index is a measure of value attached to each package of the set of all items, based on the bids at the given time.

80. A computer implemented method as recited in claim 75 wherein the subset of bids which are selected comprise all of the bids by the bidder which are in effect.

81. A computer implemented method as recited in claim 75 wherein the subset of bids which are selected comprise all of the bids by the bidder which are in effect except for provisionally-winning bids.

82. A computer system for conducting a dynamic auction for a set of items among a plurality of bidders wherein new bids are received at the computer system and an allocation of the items among the bidders is determined by the computer system based on the bids, comprising

means for receiving new bids from a bidder at the computer system, and

processing means for processing bids to determine an allocation of the items among the bidders,

wherein the means for receiving new bids includes:

means for determining, at a given time, whether said new bids should be processed by effecting a comparison between bids for items on which the bidder has bid at the given time and bids for items on which the bidder has bid at times earlier than the given time, wherein said comparison includes use of prices.

83. A computer system as recited in claim 82 wherein the means for determining includes:

means for comparing a first and a second difference, where the first difference is a difference between a minimum price for first items at the given time and a minimum price for the first items at a particular time earlier than the given time, and where the second difference is a difference between a minimum price for second items at the given time and a minimum price for the second items at the particular time earlier than the given time, and wherein

the determining means determines the new bids should be processed only if the first difference is at least as great as the second difference.

84. A computer system as recited in claim 83 wherein the first items comprise the items on which the bidder has bid at the particular time earlier than the given time, and the second items comprise the items on which the bidder has bid at the given time.

85. A computer system as recited in claim 83 wherein the first items comprise items on which the bidder has bid in combinations of bids at the particular time earlier than the given time, and wherein the second items comprise items on which the bidder has bid in combinations of bids at the given time.

86. A computer system as recited in claim 82 wherein the bids comprise package bids and the means for determining includes:

means for comparing a first and second difference, where the first difference is a difference between a minimum price for a first package of items at the given time and a minimum price for the first package at a particular time earlier than the given time, and where the second difference is a difference between a minimum price for a second package of items at the given time and a minimum price for the second package at the particular time earlier than the given time, and

the determining means determines that the new bids should be processed only if the first difference is at least as great as the second difference.

87. A computer system as recited in claim 86 wherein the first package comprises a package on which the bidder has bid at the particular time earlier than the given time, and wherein the second package comprises a package on which the bidder has bid at the given time.

88. A computer system as recited in claim 82 wherein the means for determining includes:

means for comparing a first and second difference, where the first difference is a difference between a price for first items at the given time and a price for the first items at a particular time earlier than the given time, and where the second difference is a difference between a price for second items at the given time and a price for the second items at the particular time earlier than the given time, and

the determining means determines the new bids should be processed only if the first difference is at least as great as the second difference.

89. A computer system as recited in claim 88 wherein the first items comprise the items on which the bidder has bid at the particular time earlier than the given time, and wherein the second items comprise the items on which the bidder has bid at the given time.

90. A computer system as recited in claim 88 wherein the first items comprise items on which the bidder has bid in combinations of bids at the particular time earlier than the given time, and wherein the second items comprise items on which the bidder has bid in combinations of bids at the given time.

91. A computer system as recited in claim 82 wherein the bids comprise package bids and the determining means includes:

means for comparing a first and second difference, where the first difference is a difference between a price for a first package of items at the given time and a price for the first package at a particular time earlier than the given time, and where the second difference is a difference between a price for a second package of items at the given time and a price for the second package at the particular time earlier than the given time, and

the determining means determines the new bids should be processed only if the first difference is at least as great as the second difference.

92. A computer system as recited in claim 91 wherein the first package comprises a package on which the bidder has bid at the particular time earlier than the given time, and wherein the second package comprises a package on which the bidder has bid at the given time.

93. A computer system as recited in claim 83 wherein the auction is carried out in sequential bidding opportunities and the comparing means compares said first and second differences for each bidding opportunity prior to the entering of the new bid and the determining means determines the new bids should be processed only if the first difference is at least as great as the second difference for each comparison.

94. A computer implemented method for conducting a dynamic auction for a set of items among a plurality of bidders wherein new bids are received at a computer and an allocation of the items among the bidders is determined by the computer based on bids, comprising

receiving new bids, at a computer, and

processing bids, at a computer, to determine an allocation of the items among the bidders, wherein receiving new bids from a bidder at a given time includes:

determining whether said new bids should be processed by effecting a comparison between bids for items on which the bidder has bid at the given time and bids for items on which the bidder has bid at times earlier than the given time, wherein said comparison includes use of prices, and

processing only said new bids which are indicated to be processed by the comparison.

95. A computer implemented method as recited in claim 94 wherein the comparison comprises:

comparing a first and second difference, where the first difference is a difference between a minimum price for first items at the given time and a minimum price for the first items at a particular time earlier than the given time, and where the second difference is a difference between a minimum price for second items at the given time and a minimum price for the second items at the particular time earlier than the given time, and

processing the new bids only if the first difference is at least as great as the second difference.

96. A computer implemented method as recited in claim 95 wherein the first items comprise the items on which the bidder has bid at the particular time earlier than the given time, and wherein the second items comprise the items on which the bidder has bid at the given time.

97. A computer implemented method as recited in claim 95 wherein the first items comprise items on which the bidder has bid in combinations of bids at the particular time earlier than the given time, and wherein the second items comprise items on which the bidder has bid in combinations of bids at the given time.

98. A computer implemented method as recited in claim 94 wherein the bids comprise package bids and the comparison comprises:

comparing a first and second difference, where the first difference is a difference between a minimum price for a first package of items at the given time and a minimum price for the first package at a particular time earlier than the given time, and where the second difference is a difference between a minimum price for a second package of items at the given time and a minimum price for the second package at the particular time earlier than the given time, and

processing the new bids only if the first difference is at least as great as the second difference.

99. A computer implemented method as recited in claim 98 wherein the first package comprises a package on which the bidder has bid at the particular time earlier than the given time, and wherein the second package comprises a package on which the bidder has bid at the given time.

100. A computer implemented method as recited in claim 94 wherein the comparison comprises:

comparing a first and second difference, where the first difference is a difference between a price for first items at the given time and a price for the first items at a particular time earlier than the given time, and where the second difference is a difference between a price for second items at the given time and a price for the second items at the particular time earlier than the given time, and

processing the new bids only if the first difference is at least as great as the second difference.

101. A computer implemented method as recited in claim 100 wherein the first items comprise the items on which the bidder has bid at the particular time earlier than the given time, and wherein the second items comprise the items on which the bidder has bid at the given time.

102. A computer implemented method as recited in claim 100 wherein the first items comprise items on which the bidder has bid in combinations of bids at the particular time earlier than the given time, and wherein the second items comprise items on which the bidder has bid in combinations of bids at the given time.

103. A computer implemented method as recited in claim 94 wherein the bids comprise package bids and the comparison comprises:

comparing a first and second difference, where the first difference is a difference between a price for a first package of items at the given time and a price for the first package at a particular time earlier than the given time, and where the second difference is a difference between a price for a second package of items at the given time and a price for the second package at the particular time earlier than the given time, and

processing the new bids only if the first difference is at least as great as the second difference.

104. A computer implemented method as recited in claim 103 wherein the first package comprises a package on which the bidder has bid at the particular time earlier than the given time, and wherein the second package comprises a package on which the bidder has bid at the given time.

105. A computer implemented method as recited in claim 94 wherein the auction is carried out in sequential bidding opportunities and the comparing is implemented for each bidding opportunity prior to the entering of the new bids and the new bids are processed only if the first difference is at least as great as the second difference for each comparison.

106. A computer system for conducting an auction among a plurality of bidders where the bidders use agents for generating proxy bids based on information received from a bidder, comprising:

receiving means, at the computer system, for receiving proxy bids from the agents,

processing means for processing the proxy bids to generate auction state information, and

first transmitting means for transmitting to an agent, based on particular auction state, a command to alter a change indicator variable from one state, allowing agents, during the course of an auction, to respond to inputs from bidders, to another state preventing agents, during the course of an auction, from responding to inputs from bidders.

107. A computer system as recited in claim 106 which further includes:

second transmitting means for transmitting, prior to the command, a last call command to alert bidders to an imminent transition of agents from one state, allowing agents to respond to inputs from bidders, to another state preventing agents from responding to inputs from bidders.

108. A computer system as recited in claim 107 which further includes:

third transmitting means for transmitting, prior to the last call command, a message to an auctioneer to approve or disapprove the last call command, wherein

the first and second transmitting means transmit the last call command and the change indicator value command only on approval of the auctioneer.

109. A computer system as recited in claim 106 wherein the auction is for allocating a set of items among the plurality of bidders by the computer system based on the bids, where the computer system includes:

means for determining an allocation of items to bidders based on the bids.

110. A computer system as recited in claim 109 wherein the items are dissimilar.

111. A computer system as recited in claim 109 wherein the bids are package bids.

112. A computer system as recited in claim 110 wherein the bids are package bids.

113. A computer system as recited in claim 107 wherein the auction is for allocating a set of items among the plurality of bidders by the computer system based on the bids, and the computer system includes:

means for determining an allocation of items to bidders based on the bids.

114. A computer system as recited in claim 113 wherein the items are dissimilar.

115. A computer system as recited in claim 113 wherein the bids are package bids.

116. A computer system as recited in claim 114 wherein the bids are package bids.

117. A computer implemented method for conducting an auction among a plurality of bidders where the bidders use agents for generating proxy bids based on information received from a bidder, comprising:

receiving, at a computer, proxy bids from the agents,

processing the proxy bids to generate auction state information, wherein at least one instance of auction state information results in;

transmitting a command to alter a change indicator variable from one state, allowing agents, during the course of an auction, to respond to inputs from bidders, to another state preventing agents, during the course of an auction, from responding to inputs from bidders.

118. A method as recited in claim 117 which further includes:

transmitting, prior to the command, a last call command to alert bidders to an imminent transition of agents from one state, allowing agents to respond to inputs from bidders, to another state preventing agents from responding to inputs from bidders.

119. A method as recited in claim 118 which further includes:

transmitting, prior to the last call command, a message to an auctioneer to approve or disapprove the last call command, and

conditionally transmitting the last call command and the change indicator value command only on approval of the auctioneer.

120. A method as recited in claim 117 wherein the auction is for allocating a set of items among the plurality of bidders by the computer based on the bids, which includes:

determining an allocation of items to bidders based on the bids.

121. A method as recited in claim 120 wherein the items are dissimilar.

122. A method as recited in claim 120 wherein the bids are package bids.

123. A method as recited in claim 121 wherein the bids are package bids.

124. A method as recited in claim 118 wherein the auction is for allocating a set of items among the plurality of bidders by the computer based on the bids, which includes:

determining an allocation of items to bidders based on the bids.

125. A method as recited in claim 124 wherein the items are dissimilar.

126. A method as recited in claim 124 wherein the bids are package bids.

127. A method as recited in claim 125 wherein the bids are package bids.

128. A proxy agent mediating between a bidder and an auction server in a computer implemented auction for a set of items comprising

means for receiving bid information relating to the valuation for one or more subsets of items from a bidder and auction state information from the auction server, and

means for sending bids to the auction server, where the bids sent to the auction server are directed to one or more subsets of items and are derived from the bid information.

129. A proxy agent as recited in claim 128 where the means for sending bids responds to current bid information only if responding to the current bid information is enabled by the auction state information.

130. A proxy agent as recited in claim 128 wherein the means for sending a bid selects a bid to have a price parameter for a subset which is less than or equal to the bidder'"'"'s valuation.

131. A proxy agent as recited in claim 128 wherein the means for sending a bid creates a bid by selecting a subset to maximize a quantity which is a difference between a bidder'"'"'s valuation for the selected subset and a minimum acceptable bid for the subset.

132. A proxy agent as recited in claim 128 wherein the means for sending creates a bid for a particular subset to maximize a quantity which is a difference between a bidder'"'"'s valuation and a minimum acceptable bid for that subset.

133. A proxy agent as recited in claim 132 wherein the minimum acceptable bid for a subset is determined from auction state information as a product of a constant greater than unity and an existing high bid for the set.

134. A proxy agent as recited in claim 128 wherein the means for receiving bid information receives a valuation parameter for at least one subset of items and a budget parameter.

135. A proxy agent as recited in claim 134 wherein the means for receiving bid information received bid information including a budget parameter is related to a particular subset.

136. A proxy agent as recited in claim 135 wherein the means for receiving bid information received bid information including a budget parameter which is not related to a particular subset.

137. A proxy agent as recited in claim 134 wherein the means for sending a bid creates a bid by selecting a subset to maximize a quantity which is a difference between a bidder'"'"'s valuation for the selected subset and a minimum acceptable bid for the subset, subject to a budget limitation expressed by the budget parameter.

138. A proxy agent as recited in claim 128 wherein the means for sending a bid creates a bid by selecting a subset to maximize a quantity which is a difference between a maximum acceptable bid for the selected subset and a bidder'"'"'s valuation for the subset.

139. A proxy agent as recited in claim 134 wherein the means for sending a bid creates a bid by selecting a subset to maximize a quantity which is a difference between a maximum acceptable bid for the selected subset and a bidder'"'"'s valuation for the subset, subject to a budget limitation expressed by the budget parameter.

140. A method of implementing a proxy agent mediating between a bidder and an auction server in a computer implemented auction for a set of items comprising

receiving bid information relating to the valuation for one or more subsets of items from a bidder and auction state information from a computer implemented auction server, and

sending bids to the auction server, where

the bids sent to the auction server are directed to one or more subsets of items and are derived from the bid information.

141. A method of implementing a proxy agent as recited in claim 140 where bids are based on current bid information only if responding to the current bid information is enabled by the auction state information.

142. A method as recited in claim 140 wherein a bid is selected to have a price parameter for a subset which is less than or equal to the bidder'"'"'s valuation.

143. A method as recited in claim 140 wherein a bid is created by selecting a subset to maximize a quantity which is a difference between a bidder'"'"'s valuation for the selected subset and a minimum acceptable bid for the subset.

144. A method as recited in claim 140 wherein a bid for a particular subset is selected to maximize a quantity which is a difference between a bidder'"'"'s valuation and a minimum acceptable bid for that subset.

145. A method as recited in claim 144 wherein the minimum acceptable bid for a subset is determined as a product of a constant greater than unity and an existing high bid for the subset.

146. A method as recited in claim 140 wherein the bid information includes a valuation parameter for at least one subset of items and a budget parameter.

147. A method as recited in claim 146 wherein the budget parameter is related to a particular subset.

148. A method as recited in claim 147 wherein the budget parameter is not related to a particular subset.

149. A method as recited in claim 146 wherein a bid is created by selecting a subset to maximize a quantity which is a difference between a bidder'"'"'s valuation for the selected subset and a minimum acceptable bid for the subset, subject to a budget limitation expressed by the budget parameter.

150. A method as recited in claim 140 wherein a bid is created by selecting a subset to maximize a quantity which is a difference between a maximum acceptable bid for the selected subset and a bidder'"'"'s valuation for the subset.

151. A method as recited in claim 146 wherein a bid is created by selecting a subset to maximize a quantity which is a difference between a maximum acceptable bid for the selected subset and a bidder'"'"'s valuation for the subset, subject to a budget limitation expressed by the budget parameter.

152. A computer system for conducting an auction for a plurality of items among a plurality of bidders wherein bids are received at the computer system and an allocation of the items to the bidders is determined by the computer system based on the bids, comprising:

means for receiving at the computer system a plurality of bids for the items from at least two different bidders, wherein a bid is based, at least in part, on a budget parameter, and

means for determining an allocation at the computer system based on the bids, of at least one of the items to a selected bidder.

153. A computer system as recited in claim 152 wherein the auction is a dynamic package auction, the computer system further including

means for limiting a total value of bids from the bidder, used in allocating items, in dependence on a relation between the total value of bids from the bidder and the budget parameter.

154. A computer system as recited in claim 153 wherein the computer system limits the total value of bids.

155. A computer system as recited in claim 153 wherein the computer system includes at least one agent, including

means for receiving the budget parameter, and

means for generating the bids including means for limiting a total value of the bids.

156. A computer system as recited in claim 152 wherein the auction is a dynamic package auction, the computer system further including

means for limiting a bid in dependence on a relation between the bid and the budget parameter.

157. A computer system as recited in claim 154 wherein the computer system includes at least one agent, including

means for receiving the budget parameter, and

means for generating the bids including means for limiting the bids.

158. A computer system as recited in claim 152 wherein the auction is a dynamic package auction and the auction employs proxy bidding and the computer system includes at least one agent for generating the proxy bids, said bids are proxy bids received from agents, wherein the agent includes:

means conditionally responsive to the budget parameter input from a bidder.

159. A computer system as recited in claim 158 where the agent includes:

means for limiting a bid in dependence on the budget parameter.

160. A computer implemented method for conducting an auction for a plurality of items among a plurality of bidders wherein bids are received at a computer and an allocation of the items to the bidders is determined by the computer based on the bids, comprising:

receiving, at a computer, a plurality of bids for the items from at least two different bidders, wherein a bid depends, in part, on a budget parameter, and

determining an allocation, at the computer, based on the bids, of at least one of the items to a selected bidder.

161. A method as recited in claim 160 wherein the auction is a dynamic package auction and further including

limiting a total value of bids from the bidder, used in allocating items, in dependence on the budget parameter.

162. A method as recited in claim 161 wherein the computer limits the total value of bids.

163. A method as recited in claim 161 wherein the budget parameter is input to an agent which generates the bids for the computer and wherein the agent limits the total value of the bids.

164. A method as in claim 160 wherein the auction is a dynamic package auction and further including

limiting a bid in dependence on the budget parameter.

165. A method as recited in claim 164 wherein the budget parameter is input to an agent which generates the bids for the computer and wherein the agent limits the value of the bid.

166. A method as recited in claim 160 wherein the auction is a dynamic package auction and the auction employs proxy bidding and said bids are proxy bids received from agents, wherein

said budget parameter is input to said agent by a bidder, each agent responsive to said inputs from a bidder, said method further including

conditionally making budget parameter inputs accessible to the agent.

167. A method as recited in claim 166 further including:

limiting a bid in dependence on the budget parameter.

168. A computer system for conducting a dynamic package auction of a plurality of items wherein a computer receives bids and wherein bids specify sets of items, prices and data corresponding to at least one further parameter, said system comprising:

means for receiving, at the computer, a plurality of bids for the items from at least two different bidders, and

means for determining an allocation, at the computer, based on the bids, of at least one of the items to a selected bidder in order to optimize an objective function subject to at least one selection constraint, wherein the objective function depends on the further parameter.

169. A system as recited in claim 168 wherein the objective function expresses a combination of revenues from the selected bids and the further parameter.

170. A system as recited in claim 168 wherein said further parameter is the quality of at least one of the items and said corresponding data expresses the quality of at least one of the items.

171. A system as recited in claim 168 wherein said further parameter is the terms of payment and said corresponding data expresses the terms of payment.

172. A system as recited in claim 168 wherein said further parameter is a predetermined characteristic of the bidders and said corresponding data indicates bidder identity.

173. A system as recited in claim 169 wherein said further parameter is the quality of at least one of the items and said corresponding data expresses the quality of at least one of the items.

174. A system as recited in claim 169 wherein said further parameter is the terms of payment and said corresponding data expresses the terms of payment.

175. A system as recited in claim 169 wherein said further parameter is a predetermined characteristic of the bidders and said corresponding data indicates bidder identity.

176. A computer system for conducting a dynamic package auction of a plurality of items wherein a computer receives bids and wherein bids specify sets of items, prices and data corresponding to at least one further parameter, said system comprising:

means for receiving, at the computer, a plurality of bids for the items from at least two different bidders, and

means for determining an allocation, at the computer, based on the bids, of at least one of the items to a selected bidder in order to optimize an objective function subject to at least one selection constraint, wherein the selection constraint includes the further parameter.

177. A system as recited in claim 176 wherein the objective function comprises bid revenues.

179. A system as recited in claim 176 wherein said further parameter is a quality of at least one of the items and said corresponding data expresses the quality of at least one of the items.

180. A system as recited in claim 176 wherein said further parameter is terms of payment and said corresponding data expresses the terms of payment.

181. A system as recited in claim 176 wherein said further parameter is a predetermined characteristic of the bidders and said corresponding data indicates bidder identity.

182. A system as recited in claim 176 wherein said further parameter is a quality of at least one of the items and said corresponding data expresses the quality of at least one of the items.

183. A system as recited in claim 177 wherein said further parameter is terms of payment and said corresponding data expresses the terms of payment.

184. A system as recited in claim 177 wherein said further parameter is a predetermined characteristic of the bidders and said corresponding data indicates bidder identity.

185. A computer implemented method for conducting a dynamic package auction of a plurality of items wherein a computer receives bids and wherein bids specify sets of items, prices and data corresponding to at least one further parameter, said method comprising:

receiving, at the computer, a plurality of bids for the items from at least two different bidders, and

determining an allocation, at the computer, based on the bids, of at least one of the items to a selected bidder in order to optimize an objective function subject to at least one selection constraint, wherein the objective function depends on the further parameter.

186. A method as recited in claim 185wherein the objective function expresses a combination of revenues from the selected bids and the further parameter.

187. A method as recited in claim 185wherein said further parameter is the quality of at least one of the items and said corresponding data expresses the quality of at least one of the items.

188. A method as recited in claim 185wherein said further parameter is the terms of payment and said corresponding data expresses the terms of payment.

189. A method as recited in claim 185wherein said further parameter is a predetermined characteristic of the bidders and said corresponding data indicates bidder identity.

190. A method as recited in claim 186 wherein said further parameter is the quality of at least one of the items and said corresponding data expresses the quality of at least one of the items.

191. A method as recited in claim 186 wherein said further parameter is the terms of payment and said corresponding data expresses the terms of payment.

192. A method as recited in claim 186 wherein said further parameter is a predetermined characteristic of the bidders and said corresponding data indicates bidder identity.

193. A computer implemented method for conducting a dynamic package auction of a plurality of items wherein a computer receives bids and wherein bids specify sets of items, prices and data corresponding to at least one further parameter, said method comprising:

receiving, at the computer, a plurality of bids for the items from at least two different bidders, and

determining an allocation, at the computer, based on the bids, of at least one of the items to a selected bidder in order to optimize an objective function subject to at least one selection constraint, wherein the selection constraint includes the further parameter.

194. A method as recited in claim 193 wherein the objective function comprises bid revenues.

195. A method as recited in claim 193 wherein said further parameter is a minimum quality of at least one of the items and said corresponding data expresses the quality of at least one of the items.

196. A method as recited in claim 193 wherein said further parameter is minimum terms of payment and said corresponding data expresses the terms of payment.

197. A method as recited in claim 193 wherein said further parameter is a predetermined characteristic of the bidders and said corresponding data indicates bidder identity.

198. A method as recited in claim 194 wherein said further parameter is a minimum quality of at least one of the items and said corresponding data expresses the quality of at least one of the items.

199. A method as recited in claim 194 wherein said further parameter is minimum terms of payment and said corresponding data expresses the terms of payment.

200. A method as recited in claim 194 wherein said further parameter is a predetermined characteristic of the bidders and said corresponding data indicates bidder identity.