[go: up one dir, main page]

WO2000057323A1 - Systeme et procede permettant la mise en oeuvre de techniques de vente aux encheres progressive basee sur la deuxieme offre - Google Patents

Systeme et procede permettant la mise en oeuvre de techniques de vente aux encheres progressive basee sur la deuxieme offre Download PDF

Info

Publication number
WO2000057323A1
WO2000057323A1 PCT/US1999/006384 US9906384W WO0057323A1 WO 2000057323 A1 WO2000057323 A1 WO 2000057323A1 US 9906384 W US9906384 W US 9906384W WO 0057323 A1 WO0057323 A1 WO 0057323A1
Authority
WO
WIPO (PCT)
Prior art keywords
bid
price
new
allocation
resource
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Ceased
Application number
PCT/US1999/006384
Other languages
English (en)
Inventor
Nemo Semret
Aurel Lazar
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Columbia University in the City of New York
Original Assignee
Columbia University in the City of New York
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Columbia University in the City of New York filed Critical Columbia University in the City of New York
Priority to CA002332235A priority Critical patent/CA2332235A1/fr
Priority to PCT/US1999/006384 priority patent/WO2000057323A1/fr
Priority to US09/674,024 priority patent/US7177832B1/en
Priority to JP2000607128A priority patent/JP2002540510A/ja
Publication of WO2000057323A1 publication Critical patent/WO2000057323A1/fr
Anticipated expiration legal-status Critical
Ceased legal-status Critical Current

Links

Classifications

    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q40/00—Finance; Insurance; Tax strategies; Processing of corporate or income taxes
    • G06Q40/04—Trading; Exchange, e.g. stocks, commodities, derivatives or currency exchange
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q30/00—Commerce
    • G06Q30/06—Buying, selling or leasing transactions
    • G06Q30/08—Auctions

Definitions

  • This invention relates to a system and method for performing a progressive second price auction technique in order to effectively allocate a finite resource among competing requestors.
  • a significant problem is created when a finite divisible resource is desired by two or more entities and more of the resource is requested than can possibly be allocated and distributed to the requesting entities.
  • market auctions can be created in order to properly allocate the limited resource or product to the requesting entities.
  • a finite commodity such as oil is only produced in limited quantities per month and the oil is auctioned and sold in international markets. If only one million barrels of oil is produced in a month, bidders can submit bids through the international market and typically the highest bidder will win its requested allocation of the oil and will pay the supplier of the oil the price which was bid.
  • the problem of allocating finite resources also occurs in the electronic world.
  • the Internet allows parties to communicate and exchange data over vast distances.
  • the rate of information flow can be affected by the number of users utilizing an Internet Service Provider ("ISP") and users are forced to endure long pauses in their network connections.
  • ISP Internet Service Provider
  • premium bandwidth of an ISP can be reserved for that user. While premium bandwidth can be provided by an ISP, the bandwidth is a limited resource and many users will desire all or part of the available premium bandwidth.
  • a standard auction of the bandwidth would consist of the premium bandwidth being allocated to the highest bidder for which the bidder pays the price of the bid. However, this type of auction produces inefficiencies for larger numbers of bidders and where the bids made and availability of the premium bandwidth change over time.
  • a communication network's value such as the Internet can be characterized by what economists call externalities.
  • the principles of externalities are that the value a user gets from the network depends upon the valuation and behavior of the other users.
  • One recognized principle is that a communication network is more valuable to a user if more people are connected to the network.
  • a second recognized principle is that as the utilization of the network by one user increases, the quality of service obtained by the other users decreases.
  • Resources are shared by users who because of distance, population size, or individual selfishness cannot or will not coordinate their actions sufficiently to achieve the most desirable allocation of resources.
  • This externality principles model indicates that a game theoretic approach in which other bidders actions are determinative of allocations and costs to a bidder could be used to improve the allocation of resources on a communication network.
  • the allocation of limited resources in connection with a communications network can be performed in both an actual market embodiment, when users are bidding real dollars or other items of value, and a private system embodiment, where the bids are to performed with internal budget allocations in order to obtain limited resources of the processing entity. For example, different university departments could bid on time used on a special university component with university dollars supplied by the university.
  • the requirements of an efficient pricing system include: (1) sensitivity to the range of resource requirements (either through a sufficiently broad range of traffic classes which are priced differently or by allowing users to explicitly quantify resource requirements); (2) prices must be dynamically responsive to unpredictable demand (market based system); and (3) the pricing architecture should constrain as little as possible the efficiency trade-offs. It would be desirable to achieve an auction technique which meets the above requirements and provides a more efficient allocation of a finite resource.
  • the present invention is directed to a system and method for allocating a divisible resource using a progressive second price action technique.
  • the technique retrieves data indicative of at least one previous bid comprising a quantity data component and a price data component, retrieves said data indicative of a new bid comprising a quantity component and a price component, allocates the resource to the first bids with price components higher than the new bid, allocates the resource to the new bid if there is some portion of the resource which has not yet been allocated and calculates the new bid's cost based upon the price component of the other bids which have been or would have been allocated the resource but for the new bid.
  • each bidder is encouraged to bid its true valuation for the resource and is discouraged from using inefficient bidding strategies.
  • the use of the present technique leads to a more efficient allocation system. Additional new bidders can be processed and allocated resource and the cost of the allocation to the bidder will be based upon the other bids.
  • An example of a resource which can be allocated is premium bandwidth by an Internet Service Provider (ISP).
  • the ISP can process bids from different clients requesting premium bandwidth using the second price auction technique.
  • the ISP performs an initial allocation based upon initials bids and processes each new bid after the initial allocation is done.
  • no such initial allocation is required.
  • the bids in an auction are sorted by price component, allocated resource for bids with price components above the new bid and then allocated resource for the new bid.
  • the cost of the new bid is based upon the bids of lower bidders who would have been allocated some bandwidth if the new bid had not been made. Thus the cost to the bidder is based upon others bids.
  • the cost charged to the new bid equals the sum of the quantity allocations reduced on other players weighted by the bid price of such players.
  • This technique of calculated payments or other bids encourages people to bid their actual value for the resource because the cost paid by the winning bidder is based upon other bidders' bids.
  • the ISP can also make its own floor bid in order to make sure that bids awarded allocations will pay a certain cost level.
  • Fig. 1 is a flow chart of the steps describing the progressive second price auction technique
  • Fig. 2 is a graphical example of an allocation of resources
  • Fig. 3 is another graphical example of an allocation of resources
  • Fig. 4 is a graph of a player's utility function for entering the auction as a function of possible bid prices and requested quantities;
  • Fig. 5 is a graphical example of an allocation of resources
  • Fig. 6 shows three graphs indicating the affect of a new bidder on the allocation of a resource
  • Fig. 7 shows three graphs indicating the affect of a new bidder in the auction
  • Fig. 8 shows three graphs indicating the affect of another new bidder in the auction
  • Fig. 9 shows a system of bidders and prices at the auction
  • Fig. 10 shows a diagram of a computer which could process the auction in accordance with the invention.
  • a second price auction works as follows: if an object is being auctioned between the parties, the highest bidder will be awarded the object, however, the price the winning bidder pays is the second highest price bid.
  • This type of bidding encourages bidders to bid their true valuation of the object making the auction efficient rather than attempting to engage in inefficient bidding tactics such as one dollar more than the expected highest bid. Since the highest bidder will be paying the cost of the second highest bid, there is no incentive to bid below the actual valuation of the object being bid upon. If all the other bidders are well below the bidder's bid, he will only pay the next highest bid. If the bidder bids above his valuation of the object, he may pay more than his valuation depending upon the other bidders which is not desirable. Thus there is no incentive to bid higher than valuation.
  • the progressive second price auction technique applications is extended from that of the single object example to cover bidding on divisible resources over time.
  • the technique creates a very efficient mechanism to auction and allocate a finite resource among multiple bidders. With many bidders in an auction, most players will receive a part of the resource commensurate with their valuation of the resource being auctioned.
  • Figure 1 is a flow chart of the steps for performing the progressive second price auction technique. While the technique described in Figure 1 can be used to auction any resource or other divisible good or service, the description will be of an auction of units of a generic resource. An application to an actual resources which can be allocated could be premium bandwidth for example.
  • a bid is defined as having two components: a quantity component and a price component. For example, a bidder in an auction for a resource may submit a bid for five units (quantity) at a price of $2 per unit. While only quantity and price components of a bid are described in the preferred embodiment, additional components or other information could also be included in the bid as needed.
  • Step 103 retrieves a list of bids previously made for the resource in a particular auction which have been previously sorted by decreasing bid price. This means the bid with the highest price offered will be first in the list and so on. Alternatively, the bids can be sorted after they are retrieved.
  • the list of bids can be stored in a computer file, in memory of the processor performing the auction, be inputted by the keyboard or transmitted over a network in real-time by the bidder's software bidding agreement at the time of the auction or can be retrieved by any other means when the progressive second price auction technique is performed. For example, if ten bidders have previously bid for premium bandwidth in an auction, these bids (including the quantity desired and price per unit offered) will be stored in the memory of a computer or other apparatus operating the technique.
  • the bids will then be sorted by a conventional sorting technique to place the previous bids in order from highest and lowest.
  • the bids can be stored in decreasing price order in a linked list as they are received. If the bid to be processed is the first bid made in the auction and no other bid have been stored, the bid will become the first bid in the list.
  • Step 105 retrieves the new bid (S ; ) to be processed which has at least both a quantity component and price component.
  • the quantity component is in predefined units for the auction, for example, one Mbps (Megabit per second) of bandwidth, and the price will preferably be in the form of price per unit. However, the price can also be entered in the form of total price for the quantity bid and the unit price can be easily calculated.
  • the new bidder will be encouraged to optimally submit a bid that is approximately the bidder's true valuation for the resource.
  • the bidder optimally bids his true value because of the operation of the second price auction mechanism as explained above.
  • the new bid can be submitted electronically over a network such as the Internet or can be entered manually into a computer performing the technique. For example, a new bidder desiring premium bandwidth may type in a bid of $4 per unit for three units of bandwidth into a computer and transmit the bid to the computer operating the technique over a network.
  • a software agent can be programmed to automatically bid according to the user's needs and the market as represented by the other users' bids.
  • Step 107 retrieves a total quantity (Q) of resource which is being auctioned and which is available for use at one time.
  • Q a total quantity of resource which is being auctioned and which is available for use at one time.
  • the resource being auctioned is premium bandwidth
  • the total premium bandwidth for a given ISP system or node may be 20 Mbps. Since there is only a finite amount of the premium bandwidth resource which can be used at one time, no amount of bandwidth over Q can be allocated.
  • the quantity Q can be input manually or retrieved from electronic storage or be calculated by the auction processor. Additionally Q can be a predetermined value in the technique for the resource or can be calculated based upon other parameters. For example, if the resource being allocated is bandwidth, the ISP can monitor how much premium bandwidth is available to be auctioned with respect to the total bandwidth for a given time period.
  • Step 109 initially sets the quantity remaining variable (Qj) to equal the total quantity Q.
  • Qj the quantity remaining variable
  • the quantity remaining variable Q indicates the amount of the resource which has not yet been allocated during the allocation technique for the new bid.
  • S (Q J ,P J )
  • Q j is the j th quantity bid
  • P is the j th price bid.
  • Step 113 checks if the current identified bid in the list is greater than or equal to the new bid S, in the auction. If the current identified bid in the list is greater than or equal to the new bid price, the technique continues with step 115. If the current bid pointed to in the list is less than the new bid price, than the technique continues with step 117.
  • Step 115 subtracts the quantity component of the current identified bid in the list from the value of the available quantity Q, . Since the price of the current identified bid in the list is greater than the new bid, it will be awarded the quantity of resources that it had requested (to the extent the remaining quantity of resource is not yet assigned during this allocation for the new bid). The amount of the resource which has just been allocated to the higher bid in the bid list is subtracted from the available resource variable Q, . If the new bid price is below previous bids from the bid list which in aggregate request a total allocation that equals or exceeds the available allocation of the service provider, the new bid will be allocated none of the resource because the resource has been allocated to the higher bidders. The new bidder will be forced to increase its bid price in a subsequent bid or obtain the resource from another source.
  • Step 116 checks if the quantity remaining variable Q, is equal to zero. If it is equal to zero, then there is no more resource to be allocated to the new bidder S, and the technique is completed for the new bidder S, with respect to the bid made. The bidder may then increase its bid price and submit a new bid. If Q, does not equal zero, then the testing of received bids in the list continues and the technique goes to step 111.
  • Step 117 then allocates the proper amount of the resource to the new bidder S j by taking the minimum of (1) Q, (the amount remaining) and (2) q, (the amount of requested).
  • the new bid S will be assigned the quantity requested to the extent it is available and not already assigned. In the example above, the new bidder would be assigned 3 units.
  • Steps 119 through 123 calculate the price for the new bidder according to the progressive second price auction technique.
  • Step 119 places the new bid in sorted order into the list of bids. For example, the bid could be added to a linked list such that the bid prices are arranged in descending order. The lowest bid which has previously received an allocation of the resource is then identified. If an initial allocation has not yet been performed, then the lowest bid is identified which would have been allocated some of the resource if the new bid was not present.
  • Step 121 then calculates the total cost for the new bidder based upon the progressive second price auction technique.
  • the cost is calculated on an exclusion compensation principle: the new bidder i pays for his allocation so as to cover the "social opportunity cost " which is given by the declared willingness to pay (through their bids) of the users which are excluded by i's presence. The operator and supplier of resources at the auction is thus compensated for the maximum lost potential revenue from the new allocation.
  • the stored bidding list would include the following: (3, $7), (3, $6), (3, $5), (3, $2) with the first three bids being allocated three units each.
  • the new bidder at $5.50 would be allocated the three units of the $5 bidder at a cost of $5 per unit according to the second price auction rule.
  • allocation prices are always based upon the bids of the bidder with lower bids who had previously received an allocation or would have received an allocation but for the new bidder.
  • Step 121 shows how the cost for a new bidder who is receiving an allocation of the resource is calculated.
  • the bid price is added to the total cost charged to the new bidder until the allocated quantity matches the requested quantity in the bid. This reflects the shift in allocation from the lower bidders to the new bidder with the higher bid.
  • the amount reallocated for each previous bid is responsive to how much quantity was previously allocated or would have been at that bid level.
  • the cost C is equal to C, plus the price of the unit bid times the minimum of (1) the quantity being reallocated and (2) quantity allocated for the previous bid.
  • Step 123 then identifies the next higher P j to be used in calculating C,.
  • the process then continues with step 125.
  • Step 125 checks if a, (the amount remaining be reallocated) is zero or if a P j exists which is greater than P r If one of these conditions is true, then the process is complete for the new bid. If there are additional lower P j to process and there is some resource which is to be reallocated, the process P j continues with step 121.
  • Figures 2 and 3 are graphical examples of the change in an allocation of resources in response to a new bid being processed.
  • Figure 2 depicts an allocation of resources based upon previous bids before the new bidder has entered the auction a new bid is processed.
  • the total quantity of resources (Q ⁇ ) is thirty- five units as indicated in box 203.
  • the following bids are retrieved as indicated in box 205: the first retrieved bid S, is for 7 units at $1 per unit; the second retrieved bid S 2 is for 18 units at $2 per unit; the third retrieved bid S 3 is for 5 units at $3 per unit and the fourth retrieved bid S 4 is for 10 units at $10 per unit.
  • Graph 201 shows a horizontal axis indicating bid quantity 207 and a vertical axis indicating bid price 209.
  • the bids located to the right of the vertical axis are allocated a portion of the resource being auctioned.
  • the bids are initially allocated according to the highest bid price.
  • the fourth bid S 4 (the highest bid price) is indicated by bar 211.
  • Bar 211 shows that the entire 10 units requested by the bid have been allocated.
  • Bar 213 shows the allocation for the third bid S 3 which received all of its five requested units.
  • bar 217 indicates two units being allocated to bid S, and five units requested by S, not receiving an allocation (shown to the left of the vertical axis).
  • the price levels on the graph indicate the bid price and the not price actually paid.
  • the cost of the successful bid is calculated using the progressive second price auction technique described in connection with Figure 1.
  • Figure 3 shows a graphical example of how the allocation changes if a new fifth bid comes into the auction as S 5 (5, $2.50) according to the present technique.
  • Figure 3 shows the graph 201 but now with the new allocation due to the processing of the new bid S 5 .
  • Graph 301 has a horizontal axis indicating bid quantity 307 and a vertical access indicating the bid price 309. The total level of quantity available is still 35 units as indicated in box 203.
  • the new bid S 5 is shown in the bid list 305 and is placed in the order of prices bid.
  • Bid S 5 with a bid price of $2.50 is shown between S 2 and S 3 .
  • the graph now reflects the allocation to each bid after the new bid is taken into account.
  • the entity with bid S 4 is again allocated 10 units as shown by bar 311.
  • the person with bid S 3 is again allocated 5 units as shown by bar 313.
  • the next highest bid is S 5 so the bidder of S 5 is allocated the five units requested.
  • the next highest bid, S 2 has requested an allocation of 18 units.
  • the auctioneer can submit its own floor bid to buy the resource at a particular price. This will stop initial bidders from acquiring the resources for too low a price and will allow the auctioneer to influence the price paid by others.
  • the floor price must be set at an appropriate level or the auctioneer may not sell the resources to other bidders at all.
  • the progressive second price auction technique meets the requirements of an efficient pricing system by allowing for varying bids by the bidding entity and allowing a bidder to specify the quantity requested, having resource prices which are responsive to the market conditions and limiting restraints on efficiency.
  • a valuation technique can be used to quickly enter the bids of a bidder into an electronic auction to get an efficient result.
  • Each player i has a valuation of the resource ⁇ , (a, (s)) > 0 which is the total value of the allocation received.
  • the plateaus at the top of the graph 401 show where the quantity requested by the new bidder can no longer be increased at that bid price because they have been allocated to other bidders.
  • Graph 401 has three axis: quantity 403, price bid 405 and utility 407. While the new bid price is under ten, the utility is increasing as shown in the figure. However, if the price of the new bid has a price component over 12 in order to get additional quantity, the overall utility for each extra unit decreases because the most the bidder is willing to pay for an allocation of resources is ten dollars per unit.
  • the utility graph can be shown in another form as in Figure 5.
  • Figure 5 shows an allocation graph similar to the graphs shown in Figures 2 and 3.
  • Box 501 shows the valuator function lOq (or ten times the quantity) where the quantity is shown on the horizontal axis and utility is shown in the vertical axis.
  • the horizontal axis indicates quantity allocated 505 and the vertical axis is the price of the bid 507.
  • the total quantity to be allocated is one hundred units.
  • the bids for the auction are indicated in box 509.
  • the allocation to bid S is indicated by bar 51 1 ; the allocation to bid S, is indicated by bar 513; the allocation to bid S 3 is indicated by bar 515; the allocation to bid S 4 is indicated by bar 517 and the bid S 5 is indicated by bar 519.
  • Bid S 5 is only allocated twenty units of all of the overall units.
  • the derivative of the valuation function is taken to determine the slope of the demand.
  • the derivative of the new bidder's valuation function is shown in graph 503 as line 521. Every quantity indicated in a bar below line 521 is more valuable to the new bidder than those quantities bid by other bidders with lower price components.
  • the new bidders would want to bid $10 in order to obtain seventy of the total units which includes the twenty units of line 513, the twenty units of line 515, ten units of line 517 and twenty units of line 519. Additional units could cost above the calculation for the unit so a higher bid would not be efficient.
  • the actual cost which the new bidder will pay is calculated using the technique described in connection with Figure 1.
  • the thirty units shown by bar 511 were bid at a price of $ 12 which is above the price of $ 10, the utility derived by the new bidder.
  • the new bidder would obtain 70 units at a bid price of $10. While a straight line derivative function is shown in the example, more complicated valuation functions can be used to model the correct bidding strategy.
  • auctions are applicable in a setting where the resources are arbitrarily divisible. Thus, this approach would fit well in a system where arbitrary amounts of resources can be requested and obtained, such as in pricing of the "expected capacity" profiles in the Internet with differential service.
  • bidding could be for example on a per flow basis or per unit time on an appropriate time scale.
  • a flow level model can be used particularly if the entity in the auction is a software bidding agent, possibly embedded into a software application, which develops (over time and with human feedback) an accurate estimate of the entity's utility function ⁇ , the relationship between resources and value or perceived level of service.
  • An example of a flow level auction includes an agent-based system implemented in a common programming language and a seller which is an ISP whose network has a bottleneck bandwidth (maximum resource capacity) of 45 Mbps.
  • the example described is an actual auction performed using the progressive second price auction mechanism.
  • the ISP allocates capacity to all its users.
  • the allocation is a traffic profile for each user, enforced by a profile meter at the point of connection to the network.
  • the network offers premium service to all in-profile packets (e.g. lower loss and/or delay) and any out-of-profile traffic gets best-effort service.
  • the provider must provision the network to carry all the in-profile traffic that it expects at a given time, or conversely, must not sell more capacity ("sum" of profiles) than it has in the worst case scenario.
  • the ISP will sell a total capacity which is equal or greater than the bottleneck capacity. In this example it is assumed the ISP is extremely conservative, so it will only sell 45 Mbps so that all units sold will be supplied. In addition the ISP must ensure that it in turn purchases a large enough profile from neighboring networks to ensure that the in-profile traffic of its users get the contracted service even after it leaves its network.
  • the reserve price p 0 corresponds to the price paid by the network to get capacity from neighboring networks.
  • the profiles for the bidders in this example can be any traffic descriptor.
  • the allocation can give the users a minimum guaranteed bandwidth (which will be the allocated quantity), with each user being allowed to burst up to their physical line rate. This can be done for example with a "leaky- bucket" traffic meter. If the bucket sizes are fixed by the ISP, and users are allocated (and pay for) a token rate. The bucket size determines the maximum in-profile burst size, and would correspond to the worst case buffer occupancy — or delay ⁇ that the ISP can tolerate for premium traffic. Clearly the quality of service for most internet traffic will be better if the allocated premium bandwidth is increased.
  • the user population in this example has line rates q, e ⁇ 256 Kb/s, 1.5 Mbps, 45 Mbps ⁇ .
  • ISPs currently charge Tl connections at flat rates of $500-$ 1500 per month. This works out to $0.008-$0.023 per Mbps per minute.
  • dial-up users are charged flat-rates of $10-$30 per month for 30 Kbps connections, which works out to $0.007-$0.021 per Mbps per minute.
  • the closeness between the two cases indicates that these flat rates are rough indications of average valuations.
  • the operator of the auction can also charge a small amount to process each bid.
  • the following bidding technique can be used as a strategy by an entity bidding in the progressive second price auction: if the utility for the bidder is increased by increasing the bid price or quantity requested and still remain within budget, submit a new bid with increased price and/or quantities.
  • the order of the bids is inconsequential for the bidding technique.
  • Bidders join the auction at different times and bid, for example, once per second.
  • the above described bidding technique can be described as selfish and short-sighted. It is selfish because it will submit a new bid if and only if it can improve its own utility (by more than the fee for the bid) and short-sighted because it does not take the extensive form of the auction into account, i.e., does not use strategies which may result in a temporary decrease but a better utility in the long run by tactically exploiting other users.
  • a bid fee is the fee paid to the operator of the auction to make the bid. It is typically a relatively small amount relative to the bid price.
  • the auction is run with one agent (bidding computer) per user, and the agents are distributed evenly on different workstations.
  • the auctioneer agent processing computer
  • "Arrivals" of new players correspond to Internet users whose profile was zero (pure best-effort) who have not been bidding who suddenly request a bidding profile. This could be for example users who are content with best effort traffic at night and on weekends, but require better bandwidth during business hours.
  • Figure 6 shows three graphs showing the affect of the new small user on the auction.
  • Graph 601 shows the requested quantity of bandwidth allocation for the 256 Kb/s line rate user.
  • the horizontal axis for graph 601 is time and the vertical axis is bandwidth request.
  • Graph 603 shows the allocation for the new bidder.
  • the horizontal axis for graph 603 is time and the vertical axis is bandwidth allocated.
  • Graph 605 shows the bid and paid price for the bandwidth.
  • the horizontal axis for graph 605 is time and the vertical axis is dollars per (minutes X Mbps).
  • the bid price is indicated by the solid line and the paid price is indicated by the dotted line.
  • the paid price (cost) line is lower than the bid price because of the progressive second price auction process being used.
  • the bid price and paid price are close to one another because of the relatively large number of bidders in the auction.
  • Figure 7 shows three graphs showing the affect of the new small user on the auction from a different perspective.
  • Graph 701 shows the requested quantity of bandwidth allocation for the 1.5 Mbps line rate user.
  • the horizontal axis of graph 701 is time and the vertical axis is bandwidth request.
  • Graph 703 shows the actual allocation for the new bidder.
  • the horizontal axis of graph 703 is time and the vertical axis is bandwidth allocated.
  • Graph 705 shows the bid and paid price for the bandwidth.
  • the horizontal axis of graph 705 is time and the vertical axis is dollars per (minutes X Mbps).
  • the bid price is indicated by the solid line and the paid price is indicated by the dotted line.
  • the paid price (cost) line is lower than the bid price because of the progressive second price auction process being used.
  • the bid price and paid price are close to one another because of the relatively large number of bidders in the auction.
  • Figure 8 shows three graphs showing the affect of the new small user on the auction from a different perspective.
  • Graph 801 shows the requested quantity of bandwidth allocation for the 45 Mbps line rate user.
  • the horizontal axis of graph 801 is time and the vertical axis is bandwidth request.
  • Graph 803 shows the actual allocation for the new bidder.
  • the horizontal axis of graph 803 is time and the vertical axis is bandwidth allocated.
  • Graph 805 shows the bid and paid price for the bandwidth.
  • the horizontal axis of graph 805 is time and the vertical axis is dollars per (minutes X Mbps).
  • the bid price is indicated by the solid line and the paid price is indicated by the dotted line.
  • the paid price (cost) line is lower than the bid price because of the progressive second price auction process being used.
  • the bid price and paid price are close to one another because of the relatively large number of bidders in the auction.
  • the T3 user (player 75) pays a lower price per unit than the other two.
  • This "volume discount” occurs because of the greater bandwidth allocation to player 75 and the fact that in the progressive second price auction mechanism more lower price bidders are being displaced by the higher quantity requested.
  • FIG. 9 is a diagram of an example of a system which can use the progressive second price auction technique.
  • System 901 includes Internet Service Providers ISPl 903, ISP2 905 and ISP3 907. Each Internet Service Provider is connected to a network.
  • ISPl 903 is connected to network 909
  • ISP2 905 is connected to network 91 1
  • ISP2 907 is connected to network 913.
  • Bidding entities 915 are computers which are connected to a network (either network 903, 905 and 907) and participate in auctions by sending electronic bids including a quantity request component and a bid price component to the ISP to which they are connected.
  • the bids can be individual bids or in the form of a valuation data expressed by multiple price component and quantity component bids or a valuation function expression.
  • the bidding entities 915 include a central processing unit and memory which stores a set of instructions for formulating and transmitting the bids to the auction processor (in this example the ISP).
  • An ISP can also transmit auction updates to the bidding entities 915.
  • System 901 can be used to auction premium bandwidth by an ISP.
  • the Internet is made up of a number of networks connected together such that a message will eventually be transmitted from a sending party to a receiving party. Thus if a bidder is bidding on premium bandwidth, a message will typically need to travel over numerous networks to reach its destination. Thus the premium bandwidth must be allocated by each network and the bids required to secure the overall connection contribute to the complete cost.
  • each ISP is requested by a computer which includes a processor and memory and is described in more detail in the description of Figure 10. If a bidding entity wants to transmit information to an entity that is not directly connected to the transmitting entity's ISP, the ISP needs to secure their proper bandwidth from the intervening and destination ISP. This is accomplished by the ISP becoming a bidding entity itself (as shown by box 904) where the ISP bids for another's ISP bandwidth in order to ensure that the data flow bandwidth will be allocated for the entire connection, not just through the ISP's connections.
  • ISP 903 will act as a bidding entity 904 for the auctions performed by ISP2 905 and ISP3 907 if the bandwidth from those ISPs is required by the bidders in ISP's auction.
  • An alternative arrangement is to have the bidder place bids himself in each of the ISP auctions. However, this leads to inefficiencies if the network path includes many ISPs who must allocate the bandwidth for the data exchange and the bidder is only connected to one ISP.
  • New bidding entity 917 is a bidder who has just entered the auction preferably after a number of other bids have been submitted. However, the new bidder 917 can also be the first entity to enter the auction. In the latter case there is preferably a floor bid which is provided by the ISP which is operating the auction.
  • FIG 10 is a diagram of an Internet Service Provider.
  • the ISP network bandwidth is controlled by a computer which preferable includes a central processing unit 1001, a memory 1003 connected to the CPU 1001, an input/output (I/O) port 1005 and a databus 1007 connecting processor 1001, memory 1003 and I/O port 1005.
  • the ISP preferably operates the progressive second price auction technique.
  • Program instructions implementing the progressive second price auction technique are stored in memory 1003 which are executed by processor 1001 to perform the technique as described in connection with Figure 1.
  • the bids can be stored in memory 1003.
  • the bids can be stored in a memory outside of the ISP but which is accessible to the ISP.
  • I/O port 1005 is connected to one or more networks which the Internet Service Provider manages.
  • Memory 1003 also can store data related to the allocation process such as the maximum amount of the resource which can be allocated, the present allocation of the resource and the past history of the allocation and payments made to the ISP after the bids have been accepted.
  • Processor 1001 can perform the calculation necessary to perform the allocations and calculate the costs of a new bidder using the progressive second price auction technique.
  • the auction technique is formulated in a manner which is generic enough for use in a wide range of situations.

Landscapes

  • Business, Economics & Management (AREA)
  • Accounting & Taxation (AREA)
  • Finance (AREA)
  • Engineering & Computer Science (AREA)
  • Marketing (AREA)
  • Economics (AREA)
  • Development Economics (AREA)
  • Strategic Management (AREA)
  • Physics & Mathematics (AREA)
  • General Business, Economics & Management (AREA)
  • General Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Technology Law (AREA)
  • Entrepreneurship & Innovation (AREA)
  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)

Abstract

L'invention concerne un système et un procédé pour l'attribution d'une ressource au moyen d'une technique de vente aux enchères progressive basée sur un deuxième cours acheteur. Une vente aux enchères est assurée pour une ressource limitée, telle qu'une largeur de bande d'un réseau de fournisseurs de services Internet, dans laquelle des enchères sont soumises par des utilisateurs potentiels (103), comprenant la quantité désirée et le prix pour chaque unité de ressource faisant l'objet d'une offre. Afin d'assurer une attribution efficace, un nouvel enchérisseur (105) se voit attribuer une partie de la ressource en fonction de la disponibilité de la ressource limitée (109) en raison des offres supérieures à celles du nouvel enchérisseur (113). Le prix réel payé (121) par le nouvel enchérisseur est basé sur les offres inférieures contre lesquelles la ressource aurait ou a été attribuée. Le calcul du prix payé incite les enchérisseurs à faire une offre en fonction de leur évaluation réelle de la ressource au lieu qu'ils fassent des offres tactiques inefficaces.
PCT/US1999/006384 1999-03-23 1999-03-23 Systeme et procede permettant la mise en oeuvre de techniques de vente aux encheres progressive basee sur la deuxieme offre Ceased WO2000057323A1 (fr)

Priority Applications (4)

Application Number Priority Date Filing Date Title
CA002332235A CA2332235A1 (fr) 1999-03-23 1999-03-23 Systeme et procede permettant la mise en oeuvre de techniques de vente aux encheres progressive basee sur la deuxieme offre
PCT/US1999/006384 WO2000057323A1 (fr) 1999-03-23 1999-03-23 Systeme et procede permettant la mise en oeuvre de techniques de vente aux encheres progressive basee sur la deuxieme offre
US09/674,024 US7177832B1 (en) 1999-03-23 1999-03-23 System and method for performing a progressive second price auction technique
JP2000607128A JP2002540510A (ja) 1999-03-23 1999-03-23 プログレッシブセカンドプライスオークション手法を実行するシステム及び方法

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
PCT/US1999/006384 WO2000057323A1 (fr) 1999-03-23 1999-03-23 Systeme et procede permettant la mise en oeuvre de techniques de vente aux encheres progressive basee sur la deuxieme offre

Publications (1)

Publication Number Publication Date
WO2000057323A1 true WO2000057323A1 (fr) 2000-09-28

Family

ID=22272426

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/US1999/006384 Ceased WO2000057323A1 (fr) 1999-03-23 1999-03-23 Systeme et procede permettant la mise en oeuvre de techniques de vente aux encheres progressive basee sur la deuxieme offre

Country Status (3)

Country Link
JP (1) JP2002540510A (fr)
CA (1) CA2332235A1 (fr)
WO (1) WO2000057323A1 (fr)

Cited By (10)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2002150041A (ja) * 2000-11-08 2002-05-24 Sumitomo Electric Ind Ltd オークションシステム、オークション方法
EP1510917A1 (fr) * 2003-08-27 2005-03-02 France Telecom Procédé pour allouer en temps réel une ressource entre plusieurs utilisateurs au moyen d'un mécanisme d'enchères
KR100543649B1 (ko) * 2001-02-06 2006-01-20 인터내셔널 비지네스 머신즈 코포레이션 환경 조건에 따른 컴퓨터 저장 장치의 동작 파라메터 동적 조정 방법 및 그 장치
WO2008083827A3 (fr) * 2007-01-10 2008-09-04 Fraunhofer Ges Forschung Concept pour allouer des débits de données à des fournisseurs de signaux d'information dans un réseau
US7769639B2 (en) 2002-04-15 2010-08-03 Arnaud Delenda Method and system for real-time allocation of a resource among several entities
US7788133B2 (en) 2002-04-15 2010-08-31 Arnaud Delenda Method and system for real-time allocation of a resource among several entities
US7792724B2 (en) 1999-08-25 2010-09-07 The Trustees Of Columbia University In The City Of New York System and method for allocating resources using spot market and derivative market techniques
US7853500B2 (en) 2000-04-28 2010-12-14 John Nicholas And Kristin Gross Trust U/A/D April 13, 2010 System and method for determining right of access
CN111083675A (zh) * 2019-12-03 2020-04-28 西北工业大学 工业物联网中基于社交感知的资源分配方法
US20240020709A1 (en) * 2019-12-03 2024-01-18 Climate Karma Solutions Inc. System and method for tiered pricing for scarce commodities

Families Citing this family (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US7536328B2 (en) 2003-12-30 2009-05-19 Trading Technologies International, Inc. System and method for coordinating automated and semi-automated trading tools
CN103314607A (zh) * 2011-11-08 2013-09-18 日电（中国）有限公司 用于分配反馈传输资源的方法和设备以及用于提供关于反馈传输资源值的信息的方法和设备
US8799143B1 (en) 2013-03-15 2014-08-05 Trading Technologies International, Inc Trading circles
CN105225034A (zh) * 2015-09-09 2016-01-06 南京钱宝信息传媒有限公司 交互处理方法及装置
CN105184641A (zh) * 2015-09-09 2015-12-23 南京钱宝信息传媒有限公司 交互处理方法及装置

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US5487168A (en) * 1992-06-15 1996-01-23 International Business Machines Corporation Method and system for global optimization of device allocation
US5610910A (en) * 1995-08-17 1997-03-11 Northern Telecom Limited Access to telecommunications networks in multi-service environment
US5640569A (en) * 1995-04-28 1997-06-17 Sun Microsystems, Inc. Diverse goods arbitration system and method for allocating resources in a distributed computer system
US5802502A (en) * 1993-05-24 1998-09-01 British Telecommunications Public Limited Company System for selective communication connection based on transaction pricing signals

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US5487168A (en) * 1992-06-15 1996-01-23 International Business Machines Corporation Method and system for global optimization of device allocation
US5802502A (en) * 1993-05-24 1998-09-01 British Telecommunications Public Limited Company System for selective communication connection based on transaction pricing signals
US5640569A (en) * 1995-04-28 1997-06-17 Sun Microsystems, Inc. Diverse goods arbitration system and method for allocating resources in a distributed computer system
US5610910A (en) * 1995-08-17 1997-03-11 Northern Telecom Limited Access to telecommunications networks in multi-service environment

Non-Patent Citations (3)

* Cited by examiner, † Cited by third party
Title
ANONYMOUS, "New Zealand Moves Toward Market Driven Spectrum Allocation", Spectrum Report, 01 February 1991, Vol. 1, No. 4, 1 page. *
ANONYMOUS, "Revenge of the Nerds: when government auctioneers need worldly advice, where can they turn? To mathematical economists, of course", Economist, 23 July 1994, Vol. 332, No. 7873, page 70. *
FELDMAN R.A., MEHRA R., "Auction: Theory and Applications", International Monetary Fund staff papers, September 1993, Vol. 40, No. 3, pages 485-511. *

Cited By (13)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US8131616B2 (en) 1999-08-25 2012-03-06 The Trustees Of Columbia University In The City Of New York System and method for allocating resources using spot market and derivative market techniques
US7792724B2 (en) 1999-08-25 2010-09-07 The Trustees Of Columbia University In The City Of New York System and method for allocating resources using spot market and derivative market techniques
US7853500B2 (en) 2000-04-28 2010-12-14 John Nicholas And Kristin Gross Trust U/A/D April 13, 2010 System and method for determining right of access
US8117113B2 (en) 2000-04-28 2012-02-14 John Nicholas and Kristin Gross Trust System and method for determining right of access
US8332302B1 (en) 2000-04-28 2012-12-11 Galahad Entertainment, Llc Method and apparatus for auctioning items
JP2002150041A (ja) * 2000-11-08 2002-05-24 Sumitomo Electric Ind Ltd オークションシステム、オークション方法
KR100543649B1 (ko) * 2001-02-06 2006-01-20 인터내셔널 비지네스 머신즈 코포레이션 환경 조건에 따른 컴퓨터 저장 장치의 동작 파라메터 동적 조정 방법 및 그 장치
US7769639B2 (en) 2002-04-15 2010-08-03 Arnaud Delenda Method and system for real-time allocation of a resource among several entities
US7788133B2 (en) 2002-04-15 2010-08-31 Arnaud Delenda Method and system for real-time allocation of a resource among several entities
EP1510917A1 (fr) * 2003-08-27 2005-03-02 France Telecom Procédé pour allouer en temps réel une ressource entre plusieurs utilisateurs au moyen d'un mécanisme d'enchères
WO2008083827A3 (fr) * 2007-01-10 2008-09-04 Fraunhofer Ges Forschung Concept pour allouer des débits de données à des fournisseurs de signaux d'information dans un réseau
CN111083675A (zh) * 2019-12-03 2020-04-28 西北工业大学 工业物联网中基于社交感知的资源分配方法
US20240020709A1 (en) * 2019-12-03 2024-01-18 Climate Karma Solutions Inc. System and method for tiered pricing for scarce commodities

Also Published As

Publication number Publication date
CA2332235A1 (fr) 2000-09-28
JP2002540510A (ja) 2002-11-26

Similar Documents

Publication Publication Date Title
US7177832B1 (en) System and method for performing a progressive second price auction technique
US7110977B2 (en) System and method for allocating resources using spot market and derivative market techniques
US7769639B2 (en) Method and system for real-time allocation of a resource among several entities
US20040111308A1 (en) Dynamic resource allocation platform and method for time related resources
US20060167703A1 (en) Dynamic resource allocation platform and method for time related resources
WO2000057323A1 (fr) Systeme et procede permettant la mise en oeuvre de techniques de vente aux encheres progressive basee sur la deuxieme offre
Kumar et al. A preference-based resource allocation in cloud computing systems
US20030101124A1 (en) Method and system for market based resource allocation
US20040039677A1 (en) Enhanced auction mechanism for online transactions
WO2010042324A1 (fr) Systèmes et procédés permettant de faire correspondre une ou plusieurs commandes entrantes avec une commande permanente
US20140304089A1 (en) System and Method for Conducting a Recurring Auction Using a Participant Retention Mechanism
US20140304138A1 (en) System and Method for Matching One or More Incoming Order to a Standing Order Based on Multi-Level Allocation
Dramitinos et al. An auction mechanism for allocating the bandwidth of networks to their users
Buffett et al. An algorithm for procurement in supply chain management
Lee et al. Auctions as a dynamic pricing mechanism for e-services
Teo et al. A strategy-proof pricing scheme for multiple resource type allocations
Singhal et al. A combinatorial economical double auction resource allocation model (CEDARA)
Tsai et al. Bid-proportional auction for resource allocation in capacity-constrained clouds
Gerstgrasser et al. Multi-unit bilateral trade
Andrews Understanding the effects of quota trading on mobile usage dynamics
US20150003238A1 (en) System and method for management and control of communication channels
Lee et al. A participation incentive market mechanism for allocating heterogeneous network services
Xue et al. All-pay Auction Based Profit Maximization in End-to-End Computation Offloading System
Qin Towards integration: a revenue-sharing contract in a supply chain
Gorjian Mehlabani et al. Improving virtualization and migration in combinatorial dynamic mapping for cloud services

Legal Events

Date Code Title Description
AK Designated states

Kind code of ref document: A1

Designated state(s): CA JP US

ENP Entry into the national phase

Ref document number: 2332235

Country of ref document: CA

ENP Entry into the national phase

Ref country code: JP

Ref document number: 2000 607128

Kind code of ref document: A

Format of ref document f/p: F

WWE Wipo information: entry into national phase

Ref document number: 09674024

Country of ref document: US