Overconstraint detection, rule relaxation and demand reduction in a combinatorial exchange
View Patent ↗In a combinatorial exchange, a set of rules can be input for processing of bids received in connection with the exchange. At least one bid can be received from each of a plurality of exchange participants and a desired exchange objective can be defined. A determination can be made as a function of a subset of the rules if an allocation of the bids exists that is optimal for the type of exchange being conducted. If the desired exchange objective is not satisfied, the step of determining an allocation is repeated utilizing a different subset of rules each time until the desired exchange objective is satisfied. It can be determined if the exchange is overconstrained and, if so, rules can be relaxed. Also or alternatively, the demand for a quantity of an item can be increased or decreased based on the average cost of the item.
1. A computer-implemented method of conducting a combinatorial exchange, wherein combinatorial bids allow exchange participants to bid on multiple items with a single bid, the method comprising:
(a) storing instructions in a computer storage, wherein said instructions, when executed by a processor, cause the processor to determine an allocation of bids in a combinatorial exchange;
(b) storing in the computer storage a set of rules for processing of bids received in connection with the exchange, wherein said set of rules is distinct of the instructions of step (a);
(c) storing in the computer storage from each of a plurality of exchange participants at least one bid of (1) a first bid type comprised of a quantity for each of one or more items and a single price for all the quantities of all of the items, or (2) a second bid type comprised of a first item, a quantity or a range of quantities for said first item and a unit price for said first item;
(d) storing in the computer storage a desired exchange objective;
(e) determining via a processor operating under the control of the instructions of step (a) subject to a subset of the set of rules of step (b) and subject to the desired exchange objective if an allocation of the bids exists that is optimal for the type of exchange being conducted, wherein:
if said allocation exists, it includes a subset of the bids;
if the subset of the bids includes one or more first bid type, each first bid type includes all of the items of the bid and at least part of the quantity of each item;
if the subset of the bids includes one or more each second bid type, each second bid type includes the quantity or a quantity within the range of quantities for the first item; and
(f) terminating the combinatorial exchange when the optimal allocation is determined to exist in step (e), otherwise repeat step (e) utilizing a different subset of the set of rules of step (b) for each iteration of step (e) until the optimal allocation is determined to exist.
2. The method of claim 1 , further comprising the steps of:
storing in the computer storage each item desired to be sourced during the exchange and a quantity thereof;
storing in the computer storage M ideal bids, wherein:
M is a function of at least one of (1) a minimum winners rule having the largest value, and (2) a maximum volume percentage rule having the smallest value; and
each ideal bid is for the entire quantity of each item desired to be sourced during the exchange;
and
if no allocation of the M ideal bids exists subject to the currently available set of rules, the processor designating the exchange to be overconstrained for said set or subset of rules.
3. The method of claim 2 , wherein:
M
=
#
items
min
{
1
A
,
B
2
}
+
1
where
#items=the sum of the entire quantity of each item input into the exchange;
A=the minimum winners rule having the largest value; and
B=the maximum volume percentage rule having the smallest value.
4. The method of claim 3 , wherein the method further includes rounding M up to the next whole number when the value of M includes a fraction.
5. The method of claim 3 , wherein the value of the maximum volume percentage rule having the largest value is less than or equal to the #items.
6. The method of claim 3 , wherein the currently available set of rules includes the set of rules stored in the computer storage in step (b) and any amendments, additions or deletions to said set of rules.
7. The method of claim 1 , further including the step of excluding from the determination of at least one allocation, the bid(s) of at least one bidder not having a bid included in a preceding allocation.
8. The method of claim 1 , further including the steps of:
storing in the computer storage at least one volume threshold—average cost threshold pair for at least one item of step (c);
the processor allocating a subset of the quantities of said item;
the processor dividing a sum of the bid price(s) attributable to the allocated quantities of said item by a sum of the allocated quantities of said item to obtain an average cost for said item;
in a reverse auction, if the average cost of said item is greater than the average cost threshold of one of said volume threshold−average cost threshold pairs for said item, the processor decreasing a desired quantity of said item to be sourced to less than or equal to the volume threshold of said one volume threshold−average cost threshold pair for said item; and
in a forward auction, if the average cost of said item is less than the average cost threshold of one of said volume threshold−average cost threshold pairs for said item, the processor increasing a desired quantity of said item to be sold to greater than or equal to the volume threshold of said one volume threshold−average cost threshold pair for said item.
9. The method of claim 8 , wherein:
in the reverse auction, the desired exchange objective is satisfied when the average cost for each item allocated is equal to or less than a minimum average cost threshold for said item; and
in the forward auction, the desired exchange objective is satisfied when the average cost for each item allocated is equal to or greater than a maximum average cost threshold for said item.
10. The method of claim 8 , wherein a set of volume threshold−average cost threshold pairs for at least one item is stored in the computer storage in the form of one of (1) a curve of volume threshold versus average cost threshold, (2) an algorithm wherein the volume threshold is expressed as a function of the average cost threshold, or vice versa, and (3) a set of discrete volume threshold−average cost threshold pairs.
11. A computer readable medium having stored thereon instructions which, when executed by a processor, cause the processor to perform the steps of:
(a) store in a computer storage a set of rules for processing of bids received in connection with a combinatorial exchange where combinatorial bids allow exchange participants to bid on multiple items with a single bid, wherein said set of rules is distinct of the instructions;
(b) store in the computer storage at least one bid received from each of a plurality of exchange participants, wherein each bid is comprised of: (1) at least one item, (2) an initial quantity of each item, and (3) a unit price for each item or a price for all the item(s) and their quantities;
(c) store in the computer storage a desired exchange objective;
(d) determine under the control of the instructions subject to a subset of the set of rules of step (a) and the desired exchange objective if an allocation of the bids exists that is optimal for the type of exchange being conducted, wherein if said allocation exists, it includes a subset of the bids stored in step (b) and each bid of said subset includes all of the items of the bid and at least part of the initial quantity of each item; and
(e) terminate the combinatorial exchange when the optimal allocation is determined to exist in step (d), otherwise repeat step (d) utilizing a different subset of the set of rules of step (a) for each iteration of step (d) until the optimal allocation is determined to exist.
12. The computer readable medium of claim 11 , wherein the instructions further cause the processor to:
receive each item desired to be sourced during the exchange and a quantity thereof;
receive M ideal bids, wherein:
M is a function of at least one of (1) a minimum winners rule having the largest value, and (2) a maximum volume percentage rule having the smallest value; and
each ideal bid is for the entire quantity of each item desired to be sourced during the exchange;
and
if no allocation of the M ideal bids exists subject to the currently available set of rules, designate the exchange to be overconstrained for said set of rules.
13. The computer readable medium of claim 12 , wherein:
M
=
#
items
min
{
1
A
,
B
2
}
+
1
where
#items=the sum of the entire quantity of each item input into the exchange;
A=the minimum winners rule having the largest value; and
B=the maximum volume percentage rule having the smallest value.
14. The computer readable medium of claim 12 , wherein the instructions further cause the processor to round M up to the next whole number when the value of M includes a fraction.
15. The computer readable medium of claim 12 , wherein the value of the maximum volume percentage rule having the largest value is less than or equal to the #items.
16. The computer readable medium of claim 12 , wherein the currently available set of rules includes the set of rules input in step (a) and any amendments, additions or deletions to said set of rules.
17. The computer readable medium of claim 11 , wherein the instructions further cause the processor to exclude from the determination of at least one allocation, the bid(s) of at least one bidder not having a bid included in a preceding allocation.
18. The computer readable medium of claim 11 , wherein, the instructions further cause the processor to:
associate at least one volume threshold−average cost threshold pair with at least one item of step (b);
allocate a subset of the quantities of said item;
divide a sum of the bid price(s) attributable to the allocated quantities of said item by a sum of the allocated quantities of said item to obtain an average cost for said item;
in a reverse auction, if the average cost of said item is greater than the average cost threshold of one of said volume threshold−average cost threshold pairs for said item, decrease a desired quantity of said item to be sourced to less than or equal to the volume threshold of said one volume threshold−average cost threshold pair for said item; and
in a forward auction, if the average cost of said item is less than the average cost threshold of one of said volume threshold−average cost threshold pairs for said item, increase a desired quantity of said item to be sold to greater than or equal to the volume threshold of said one volume threshold−average cost threshold pair for said item.
19. The computer readable medium of claim 18 , wherein:
in the reverse auction, the desired exchange objective is satisfied when the average cost for each item allocated is equal to or less than a minimum average cost threshold for said item; and
in the forward auction, the desired exchange objective is satisfied when the average cost for each item allocated is equal to or greater than a maximum average cost threshold for said item.
20. The computer readable medium of claim 18 , wherein the plurality of volume threshold−average cost threshold pairs is stored in the form of one of (1) a curve of volume threshold versus average cost threshold, (2) an algorithm wherein the volume threshold is expressed as a function of the average cost threshold, or vice versa, and (3) a set of discrete volume threshold−average cost threshold pairs.