GB2432232A - A method of purchasing goods and/or services - Google Patents

A method of purchasing goods and/or services Download PDF

Info

Publication number
GB2432232A
GB2432232A GB0521028A GB0521028A GB2432232A GB 2432232 A GB2432232 A GB 2432232A GB 0521028 A GB0521028 A GB 0521028A GB 0521028 A GB0521028 A GB 0521028A GB 2432232 A GB2432232 A GB 2432232A
Authority
GB
Grant status
Application
Patent type
Prior art keywords
lt
gt
purchasers
plurality
group
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.)
Withdrawn
Application number
GB0521028A
Other versions
GB0521028D0 (en )
Inventor
Robert Cole
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.)
BUNDLES Ltd
Original Assignee
Bundles Ltd
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

Links

Classifications

    • GPHYSICS
    • G06COMPUTING; CALCULATING; COUNTING
    • G06QDATA PROCESSING SYSTEMS OR METHODS, SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL, SUPERVISORY OR FORECASTING PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL, SUPERVISORY OR FORECASTING PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q40/00Finance; Insurance; Tax strategies; Processing of corporate or income taxes
    • G06Q40/04Exchange, e.g. stocks, commodities, derivatives or currency exchange
    • GPHYSICS
    • G06COMPUTING; CALCULATING; COUNTING
    • G06QDATA PROCESSING SYSTEMS OR METHODS, SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL, SUPERVISORY OR FORECASTING PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL, SUPERVISORY OR FORECASTING PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q30/00Commerce, e.g. shopping or e-commerce
    • G06Q30/06Buying, selling or leasing transactions
    • G06Q30/08Auctions, matching or brokerage
    • GPHYSICS
    • G06COMPUTING; CALCULATING; COUNTING
    • G06QDATA PROCESSING SYSTEMS OR METHODS, SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL, SUPERVISORY OR FORECASTING PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL, SUPERVISORY OR FORECASTING PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q50/00Systems or methods specially adapted for specific business sectors, e.g. utilities or tourism
    • G06Q50/10Services
    • G06Q50/16Real estate

Abstract

A method of purchasing goods or services is disclosed, the method being particularly relevant to the purchase of mortgages. The method includes grouping a plurality of purchasers into groups according to characteristics relating the purchasers by using a solution estimator for an NP-complete optimisation problem. Seeking a seller to provide the required goods or services to the groups of purchasers typically by a Dutch auction conducted over the internet.

Description

<p>A method of purchasing of goods and/or services The present invention

relates to a method for purchasing goods and/or services and relates particularly, but not exclusively, to a method of purchasing goods and services in an internet auction.</p>

<p>It is well known that if a number of purchasers, who are all looking to buy identical or similar products or services, join together to form a group who collectively seek a seller, that group of purchasers will almost always be able to buy the goods or services at a cheaper price than if the purchasers approached the seller individually. An example of such a purchasing method is disclosed in an International Patent Application published under number WOO1/l3300, the disclosure in which is wholly incorporated herein by reference. In this method, a "demand collection engine" is used to collect potential purchasers together and these purchasers are grouped dependent upon a number of characteristics and these grouped purchasers offered in an internet-based Dutch auction.</p>

<p>However, such a method can only be used in simple and straightforward examples with a limited number of characteristics since the complexity of grouping the purchasers together increases significantly as the number of characteristics increase. Therefore for certain types of purchases, the factors or characteristics which distinguish one purchaser from another makes the grouping of the purchasers together into groups extremely complex because the purchasers within these groups must be sufficiently similar to each other for a seller to offer a group purchase discount.</p>

<p>Furthermore, some purchasers are easily grouped together into reasonably sized groups, for example those who have the most commonplace requirements. However, purchasers with less commonplace characteristics are therefore difficult to aggregate together into a group which will be sufficiently attractive to a potential seller to be offered a group purchase discount. It is therefore typically expected that purchasers with unusual purchase requirements will not be able to take advantage of group purchase discounts.</p>

<p>Preferred embodiments of the present invention seek to overcome disadvantages of the prior art, including, but not limited to, those set out above.</p>

<p>According to an aspect of the present invention there is provided a method of purchasing goods and/or services, the method comprising the steps of:-receiving first data collected from a plurality of potential purchasers, said first data representing a plurality of characteristics relating to said purchasers and/or the goods and/or services to be purchased; grouping a plurality of said purchasers into a plurality of first groups according to said characteristics using a solution estimator for an NP-complete optimisation problem to determine into which of said first groups a purchaser should be grouped; and outputting second data in order to seek at least one seller to provide said goods and/or services to at least one said first group.</p>

<p>By providing a method of purchasing goods and/or services in which characteristics relating to the purchaser or goods or services are collected, these purchasers are grouped together using a solution estimator for a NP-complete problem and data is output in order to find sellers for those grouped purchasers, the advantage is provided that for goods or services having a significant number of characteristics which distinguish one purchaser from another, the purchasers can still be grouped together in such a way as to offer groups whose constituent purchasers are sufficiently similar to attract sellers to offer a significant group discount.</p>

<p>Furthermore, purchasers whose characteristics are less commonplace can also be successfully grouped together to obtain a group discount from a seller.</p>

<p>The method may further comprise collecting said first data.</p>

<p>The method may also further comprise seeking at least one seller to provide said goods and/or services to at least one said first group.</p>

<p>In a preferred embodiment at least one said seller is sought by means of an auction.</p>

<p>In another preferred embodiment said auction is a dutch auction.</p>

<p>In a further preferred embodiment said grouping of purchasers into a plurality of first group comprises:-a) grouping a plurality of said purchasers in a first plurality of second groups according to first predetermined values of at least one first characteristic; and b) selecting purchasers from at least one said second group according to predetermined values of at least one second characteristic to define at least one said first group.</p>

<p>The method may further comprise:-c) grouping a plurality of said purchasers into a second plurality of second groups according to second predetermined values of at least one said first characteristic, different from at least one said first predetermined values of at least one said first characteristic; and d) selecting purchasers from at least one said second group according to predetermined values of at least one second characteristic to define at least one said first group.</p>

<p>The method may further comprise repeating said steps (c) and (d) until all said purchasers have been grouped into said first groups or no further purchasers can be selected from said at least one said second group according to predetermined values of said at least one second characteristic.</p>

<p>In the above process the purchasers are grouped together into a set of second groups or preliminary groups defined by a first predetermined set of first characteristics of the purchasers. Typically the first characteristics against which the first predetermined values are applied to determine which second group the purchaser is placed would be all the characteristics except one, which is the second characteristic.</p>

<p>From within those second groups purchasers are selected by using a second characteristic of the purchasers, that is a rule is applied to that second characteristic of those purchasers to select them into a first group from where they can be offered to sellers. The remaining purchasers are then reallocated into the second groups by using a second set of first predetermined values. The process of selecting the purchaser from the second groups is then repeated to form further first groups for offering to sellers. These steps of reallocation of purchasers into the second groups with further second sets of first predetermined values are repeated until all or almost all purchasers have been grouped into first groups. As a result, the advantage is provided that purchasers who have unusual characteristics can still be grouped together within groups whose characteristics are still sufficiently similar to be attractive to potential sellers since a second set of predetermined values of the first characteristic are chosen so as to remain attractive whilst still obtaining the best possible price for groups with more commonplace characteristics.</p>

<p>In a preferred embodiment purchasers are selected from said at least one said second group according to said predetermined values by starting with either the highest or lowest value of said second characteristic.</p>

<p>According to another aspect of the present invention there is provided a method of purchasing goods and/or services, the method comprising the steps of:-receiving first data collected from a plurality of potential purchasers, said first data representing a plurality of characteristics relating to said purchasers and/or the goods and/or services to be purchased; grouping a plurality of said purchasers into a plurality of first groups by, a) grouping a plurality of said purchasers in a first plurality of second groups according to first predetermined values of at least one first characteristic, and b) selecting purchasers from at least one said second group according to predetermined values of at least one second characteristic to define at least one said first group; and outputting second data in order to seek at least one seller to provide said goods and/or services to at least one said first group.</p>

<p>The method may further comprise:-c) grouping a plurality of said purchasers into a second plurality of second groups according to second predetermined values of at least one said first characteristic, different from at least one said first predetermined values of at least one said first characteristic; and d) selecting purchasers from at least one said second group according to predetermined values of at least one second characteristic to define at least one said first group.</p>

<p>The method may also further comprise repeating said steps (c) and (d) until all said purchasers have been grouped into said first groups or no further purchasers can be selected from said at least one said second group according to predetermined values of said at least one second characteristic.</p>

<p>In a preferred embodiment purchasers are selected from said at least one said second group according to said predetermined values by starting with either the highest or lowest value of said second characteristic.</p>

<p>According to a further aspect of the present invention there is provided an apparatus for purchasing goods and/or services, the apparatus comprising at least one processor for:-receiving first data collected from a plurality of potential purchasers, said first data representing a plurality of characteristics relating to said purchasers and/or the goods and/or services to be purchased; grouping a plurality of said purchasers into a plurality of first groups according to said characteristics using a solution estimator for an NP-complete problem to determine into which of said first groups a purchaser should be grouped; and outputting second data in order to seek at least one seller to provide said goods and/or services to at least one said first group.</p>

<p>According to another aspect of the present invention there is provided a data structure for use by a computer for purchasing goods and/or services, the data structure comprising: -first computer code executable to receive first data collected from a Plurality of potential purchasers, said first data representing a Plurality of characteristics re'ating to said Purchasers and/or the goods and/or services to be purchased; second computer code executable to group a Plurality of said purchasers into a plurality of first groups according to said characteristics using a solution estimator for an NP-complete problem to determine into which of said first groups a purchaser should be grouped; and third computer code executable to output second data in order to seek at least one seller to provide said goods and/or services to at least one said first group.</p>

<p>According to a further aspect of the present invention there is provided a computer running a program as defined above.</p>

<p>According to another aspect of the present invention there is provided a data carrier having a program as defined above.</p>

<p>Preferred embodiments of the present invention will now be described, by way of example only, and not in any limitative sense, with reference to the accompanying drawings in which:-Figure 1 is an overview of a system architecture for Performing the method of the present invention; Figure 2 is a flow chart of an embodiment of the present invention; Figure 3 is a flow chart of an embodiment of the grouping process of the present invention; Figure 4 is a flow chart showing the grouping of purchasers in an embodiment of the present invention; Figures 5a and 5b are a flow chart showing the steps involved in the combination of groups in an embodiment of the present invention; and Figures 6 -16 show a simplified worked example using the method of the present invention Referring to Figure 1, a method of Purchasing goods and/or services, in the example described herein mortgage services, a plurality of borrowers or purchasers io access web servers 12 via the internet, and provide information relating to the goods they wish to purchase, the purchasers themselves and/or other characteristics relevant to the purchase. This information is stored on a database 14 and the purchasers grouped together and offered for auction on an auction system 16 to potential sellers 18, being lenders in the case of mortgages.</p>

<p>Referring to Figure 2, a borrower 10 enters data relating to the purchase, the goods and/or services being purchased or other characteristics relevant to the purchase at step 20.</p>

<p>This data is entered via a website and saved on a web server 12. If the borrower 10 decides to get a proposal from an auction of potential borrowers, a request is submitted at step 22. when the information has been gathered from a number of borrowers io, the data, which represents a plurality of characteristics, is sorted using values of these characteristics into a plurality of second groups or bins at step 24.</p>

<p>A bundling algorithm is run, at step 26, to select purchasers from the second groups and place them into first groups. This is done by sweeping through the second groups or bins, at step 28, using a single bin algorithm to select purchasers from the bins using a rule which sorts the purchasers from the second groups or bundles into first groups having similar characteristics, step 30. In the case of mortgages, the similarity of the characteristics mean that the purchasers have similar risk to mortgage lenders of defaulting on the mortgage. These first groups or bundles of purchasers are then offered to sellers 18 or mortgage providers via an internet auction.</p>

<p>The remaining purchasers in the second groups or bins who were not selected to be put into the bundles remain with in the depleted bins 32. A combination algorithm is run at step 34 to combine, within a set of rules, the depleted bins or depleted second groups so as to form new or combined second groups/bins 36. The sweeping process 28 is again applied to these combined bins to form more first groups 30 and further depleted second groups 32. If the combination algorithm has not been run for each possible characteristic step 38, the combination process 34 is repeated until all allowable combinations of bins has been completed and the process is finished at step 40.</p>

<p>In step 26 the second groups or bins are defined by taking each purchaser and selecting the second group or bin into which they will be placed dependent upon the characteristics in the first data. In the case of mortgage purchasers, these characteristics can include credit status, employment status, the purpose of the mortgage, the loan to value ratio (LTv) and data relating to the loan amount is also collected. The second group or bin into which the purchaser is placed is dependent upon their credit status, employment status, mortgage purpose and loan to value ratio with the only Potentially variable factor within each second group being the loan value. As a result, when the potential purchasers are first put within the second groups, the characteristics of each of those purchasers is the same for all of the characteristics except the loan value. it should be pointed out that the loan value of two purchasers can be the same and for some characteristics, such as loan to value ratio, the characteristic 5 defined as the loan to value ratio being within a range whereas the actual loan to value ratios of two purchases within the same range may well be different.</p>

<p>Figure 3 shows the single bin algorithm 28 in greater detail. As indicated at 42, this process is run for each of the second groups. The purchasers within each second group are sorted, at step 44, by the size of the loan. There are two potential selection methods which can be used to extract potential purchasers from the second groups to put them into the first groups, either by starting with the largest loan and grouping together all]oan amounts that are not less than, for example, 75% of the largest loan, or alternatively starting with the smallest loan and including all of those not more than 75% larger than the smallest loan size.</p>

<p>These alternative techniques are used so that any purchasers who are not located into a first group are likely to be located towards the median of the second group they are left in. As a result, the more difficult to group together purchasers, those with less commonplace loan value requirements, are grouped together as soon as possible. At this early stage it is easier to form first groups or bundles because the population within the second groups or bins is larger. Furthermore, the remaining purchasers will tend to be closer to the median value of the population and therefore, in spite of the smaller population, should remain relatively easy to form into first groups.</p>

<p>The decision as to whether to start the selection process using the largest or the smallest loan is taken at step 46 and in this example, is decided as a result of whether the largest loan is greater than 350,00Q. If the loan size is greater than this figure then the process is started with the largest loan at step 48, if it is less than this figure, then the process starts with the smallest loan, at step 50.</p>

<p>Starting with the largest loan, at step 48, the next largest loan size is grouped with the largest at step 52 as a temporary first group. If this temporary first group is a sufficient size to qualify as a first group, step 54, then the temporary first group is saved as a full first group or bundle, step 56, and is ready for auction. If the required group size has not been achieved, the next largest loan is checked to determine whether it is less than 75% of the largest loan size, step 58. If it is not, then the next purchaser is added to the temporary first group at step 52 and the group size check, step 54, is repeated until the required first group size is reached.</p>

<p>If, before the required group size is reached, at step 58 the loan amount of the next purchaser is less than 75% of the largest loan request and the bundle is too small, step 60, then a first group cannot be formed and the remaining mortgage applications are left in the second group at step 62. If at step 60 the temporary first group size is not too small then the temporary first group is saved as a permanent first group at step 56.</p>

<p>Once a permanent first group has been successfully formed at step 56, the process is restarted using the smallest loan at step 50. The next smallest loan is grouped with the smallest at step 64 as a temporary first group and if the required bundle size is achieved at step 66 the bundle is saved in preparation for auction at step 68. If the bundle size is not achieved, the next smallest loan application is checked to see whether the smallest loan size is less than 75% of this next loan application at step 70. If it is not, this next loan request is added at step 64 and the process repeated. If the smallest loan is less than 75% of the next loan application, and the bundle size is too small as checked at step 72, then the application finishes at step 62. If a first group is successfully formed at step 68 then the process restarts with the largest remaining loan at step 48. This process is repeated with the largest and smallest loan sizes until no more first bundles can be formed and the application finishes at step 62.</p>

<p>Referring to Figure 4, once an initial sweep of the second groups or bins has been completed to form first groups or bundles, the remaining potential purchasers need to be grouped into further first groups so as to he offered to potential sellers. This is done by reconfiguring the second groups and then rerunning the single bin algorithm to select purchasers for the first groups. This reconfiguration of the second groups can either be seen as combining the existing second groups with the depleted number of potential purchasers therein after the removal of some purchasers to form the first groups. This combination of the second groups occurs within a set of predetermined rules as to which second groups can be combined together. Alternatively, as the reconfiguration of the second groups can be regarded as redefined using a different set of rules than the initial definition to repopulate the second groups. This second set of rules being different from the first rules which were used to populate the second groups in the first run and having the same net effect of combining some of the depleted groups.</p>

<p>In Figure 4 it can be seen that the process is run on a trial basis for different possible combinations of the second groups.</p>

<p>After the depleted second groups are identified at step 74, a trial combination of the groups or trial reallocation of purchasers within the new second groups or bins is performed at step 76. A trial run of selecting purchasers from the second groups to place into first groups, as set out in Figure 3, is performed at step 78 and the number of first groups which can be formed recorded. If, at step 80, all possible combinations of depleted second groups or reallocation of purchasers into new second groups has not been tried, the process returns to step 76 and a new combination or new reallocation is performed.</p>

<p>This process is repeated until all of the combinations have been tried and at step 82, the combination of second groups or reallocation of purchasers into second groups which are then selected using the process set out in Figure 3 to form first groups. At step 84 this leaves further remaining purchasers within the further depleted second groups. The process is then repeated from step 74 for these further depleted second groups.</p>

<p>The characteristics of the purchaser or goods and/or services they wish to purchase can be used to form a tree of the type shown in Figures 6 -16. Each characteristic resulting in two or more branches and each of these branches leading to further branches for further characteristics. At the end of each of these branches are the second groups or bins into which the purchasers are located if they have characteristics which follow along each of the branches which leads to that leaf. Each characteristic which leads to a branching or node in the tree is a level n with the first characteristic forming the root of the tree having level 1 and the leaves at the other end having the highest level N. The leaves being equivalent to the bins or second groups.</p>

<p>Below set out in table 1 is an example of the allowable combinations of loan to value ratio. When a purchaser enters information relating to their mortgage application the value of the property compared to the amount of money borrowed can be calculated as a ratio. This ratio can then define a characteristic by being allocated into, for example, one of nine ranges. When it comes to combining the second groups together, certain combinations of the ranges of loan to value ratio are allowed and these are set out in table 1. Where an "X" is shown the loan to value range indicated in that column can be combined with the loan to value ratio in that row. It can therefore be seen referring to column D, loan to value ratios of 71 to 75, that this ratio could be combinod with loan to value ratios of 66 to 70, row C, and/or 76 to 80, row E. It should be noted that the effect of combining, for example, column C with row B is the same as combining column B with row C.</p>

<p>A B C D E F G H I</p>

<p><60 61-65 66-70 71-75 76-80 81-85 86-90 91-95 96+ A <60 B 61-65 X X C 66-70 X X _D 71-75 X X E 76-80 ______ ______ _____ -X X F_ 81-85 X X G 86-90 X X H 91-95 ____ ____ --____ ____ X I 96+ X</p>

<p>Table 1</p>

<p>It can therefore be seen that there are a number of different ways in which the nine loan to value ratio ranges can be combined within the allowable rules set out in Table 1. In the present invention it is therefore advantageous to try a number of these possible combinations to find which provides the greatest number of sellable first groups or bundles. Some of these possible combinations are set out in Table 2. Taking the simplest example of the first two rows in table 2, one possible set of combinations, set out in row 1, is to combine range A (less than 60) with range B (61-65 to combine ranges C and D, B and F, G and H, leaving range I uncombined) Alternatively, in row 2 range A can remain uncombined whilst ranges B and C, D and B, F and G, H and I are combined together. Further possible combinations are set out in rows 3 to 5 in table 2. Other combinations of ranges are also possible although these are not set out in Table 2.</p>

<p>Row Combinations 1 A+B, C+D, E+F, G+H, I 2 A, B+C, D+E, F+G, H+I 3 A+B, C+D+E, F+G+H, I 4 A, B+C+D, E+F+G, H+I A+B+C, D+E+F, G+H+I</p>

<p>Table 2</p>

<p>The combination of second groups or redistribution of purchasers within the second group starts at step 86. For the first combination of second groups, this starts with the leaves at the end of the tree and the next node is selected at step 88. For the purposes of this example this node is the loan to value ratios. At step 90 the first row of allowable combinations is taken from Table 2 and the remaining applicants in the second groups are combined so that a second group with loan to value ratios in column A of Table 1 (less than 60) are combined with loan to value ratios in row B (61 to 65)at step 92. The single bin algorithm set out in Figure 3 is run at step 94 and the number of first groups created and purchasers who were not grouped into first groups recorded. The next set of potential combinations in the current row on Table 2 is selected at step 98 and combined at step 92. This process is repeated until at step 100 all of the bins or sets in the current row of Table 2 have been combined. Once all of these combinations in that row have been completed, the total number of first groups and remaining ungrouped purchasers are added together and recorded. At step 104 the next row in Table 2 is selected and this process repeated for the allowed combinations starting at step 92. At step 106 once all the rows in Table 2 have been completed, the row of Table 2 which has produced the most groups or bundles is selected at step 108 and the combination process of steps 92 to 100 is repeated at steps 110 to 118 with the addition of step 120 in which the first groups are formed and the data exported in order that these first groups can be offered for sale by auction.</p>

<p>This process is repeated for all of the nodes on that level of the tree at step 122 and then the process is further repeated by selecting the next level of nodes, that is the next set of characteristics, step 124, until all of the levels or characteristics have been combined where possible and the process is then finished at step 128. Any purchasers who have not been successfully allocated to a first group or bundle can be held over until the process is restarted with a new set of potential purchasers.</p>

<p>In Figures 6 to 16, a very simplified example of the method of the present invention is run through using a significantly reduced number of purchasers, namely only 10, and using only five characteristics. The characteristics are used to create the tree thereby having five levels. At level one, the route of the tree, the credit status of the purchasers is indicated. In this instance, all ten purchasers have good credit status. At level 2, the employment status of the purchasers creates the first branching of the tree into an employed branch 2a and self-employed branch 2B, splitting at a node 2c. At level 3, which is the criteria relating to the purpose of the mortgage, the employed branch splits again into a mortgage branch 3a and a remortgage branch 3b with the split at node 3c. The self-employed branch 2b does not split and runs directly into a remortgage branch 3d.</p>

<p>Where the loan to value ratio criteria is introduced, at level 4, with ranges set out in Table 1, further branches are produced. At the end of each branch a bin, or second group, 4a to 4f contains each of the ten purchasers 5a to 5j.</p> <p>It can therefore be seen that within bin 4a two purchasers 5a and 5b

have been placed, these purchasers looking for mortgages with differing loan values of 120,000 and 1,500,000.</p>

<p>Referring to Figure 7, when the single bin algorithm, set out in Figure 3, is applied to these bins 4a to 4f, only two of the purchasers within any one of the bins, namely 5d and 5e, have loan values within 75% of each other. As a result, the only first group, or bundle, 6a which can be formed from any of the second groups 4a to 4f comes from bin 4c and contains purchasers 5d and 5e. In this example the bundle size has been restricted to two purchasers whereas normally it would be considerably larger.</p>

<p>In Figure 8, the remaining purchasers can be seen in bins 4a to 4f. Since no more bundles can be formed, the second groups 4a to 4f must be reconfigured to create new second groups or bins so that further first groups or bundles can be formed. By looking at the branches in level 4, the loan to value ratio, it can be seen that the bins 4a and 4b cannot be combined because the loan to value ratios of 71 to 75 and greater than 96 are not an allowable combination as set out in column D and row I of Table 1. However from bins 4c, 4d and 4e, all of which extend from branch 3b, bins 4c and Id can be combined since their loan to value ratios are 66 to 70 and 71 to 75 respectively, see column C, row 0 in Table 1.</p>

<p>In Figure 9, a new bin 4c+d is formed containing purchasers 5f, 5g and 5h. When the single bin algorithm is applied to bin 4c+d, the purchasers 5f and 5h can be grouped into a bundle 6b.</p>

<p>The remaining purchasers 5a, 5b, Sc, 5g, Si and Sj can be seen in Figure TO and it can be seen that if the single bin algorithm were applied to these bins, no further bundles could be formed.</p>

<p>The process of combining bins then moves to the level 3 characteristic, mortgage purpose, and allows the combination of the mortgage and remortgage branches 3a and 3b which in turn results in the joining of bins 4a and 4c+d to form a new bin 4a+c+d since the loan to value ratios of the purchasers 5a, 5b and 5g are all within the range 71 to 75. When the single bin algorithm is applied to bin 4a+c+d, a new bundle 6c, containing purchasers 5a and 5g can be formed.</p>

<p>Referring to Figure 12, by rerunning the level 4 bin combinations, it can be seen that the bins 4b and 4e can be combined since the loan to value ratios of these bins, namely greater than 96 and 91 to 95 respectively, can effectively be combined according to Table 1, then a new bin 4b+e containing purchasers 5c and Se can be formed. When the single bin algorithm is applied to bin 4b+e, a new bundle 6d can be formed.</p>

<p>The remaining two bins, shown on Figure 13, namely 4a+c+d and 4f contain purchasers 5b and 5j respectively. All of the combinations of bins at level 4 and level 3 have been completed and therefore the level 2 criteria, employment status is reviewed for possible combination. According to the predefjned rule, it is allowable to combine purchasers having employed and self-employed service and this initial combination is shown in Figure 14. As we have already seen, it is possible to combine the level 3 characteristic of mortgage purpose for mortgages and remortgages and this combination is shown on Figure 15. At category level 4, the loan to value ratio of 66 to 70 and 71 to 75 can be combined according to the rules set out in table 1, and therefore a new bin, as shown on Figure 16, 4a+c+d+f contains the two remaining purchasers Sb and 5j. Since these purchasers are looking to borrow amounts of money within 75% of each other, these purchasers can be placed into a bundle 6e.</p>

<p>In order to produce the bundles with the least undesirable risk, the order in which the criteria are allowed to be combined is important. The combinations which are likely to result in the greatest difference in risk between purchasers should be applied later whilst more acceptable risk combinations applied earlier.</p>

<p>It will be appreciated by persons skilled in the art that the above embodiment has been described by way of example only, and not in any limitative sense, and that various alterations and modifications are possible without departure from the scope of the invention as defined by the appended claims.</p>

Claims (1)

  1. <p>Claims 1. A method of purchasing goods and/or services, the method
    comprising the steps of:-receiving first data collected from a plurality of potential purchasers, said first data representing a plurality of characteristics relating to said purchasers and/or the goods and/or services to be purchased; grouping a plurality of said purchasers into a plurality of first groups according to said characteristics using a solution estimator for an NP-complete optimisation problem to determine into which of said first groups a purchaser should be grouped; and outputting second data in order to seek at least one seller to provide said goods and/or services to at least one said first group.</p>
    <p>2. A method according to claim 1, further comprising collecting said first data.</p>
    <p>3. A method according to claim 1 or 2, further comprising seeking at least one seller to provide said goods and/or services to at least one said first group.</p>
    <p>4. method according to claim 3, wherein at least one said seller is sought by means of an auction.</p>
    <p>5. A method according to claims 4, wherein said auction is a dutch auction.</p>
    <p>6. A method according to any one of the preceding claims, wherein said grouping of purchasers into a plurality of first group comprises:-a) grouping a plurality of said purchasers in a first plurality of second groups according to first predetermined values of at least one first characteristic; and b) selecting purchasers from at least one said second group according to predetermined values of at least one second characteristic to define at least one said first group.</p>
    <p>7. A method according to claim 6, further comprising:-c) grouping a plurality of said purchasers into a second plurality of second groups according to second predetermined values of at least one said first characteristic, different from at least one said first predetermined values of at least one said first characteristic; and d) selecting purchasers from at least one said second group according to predetermined values of at least one second characteristic to define at least one said first group.</p>
    <p>8. A method according to claim 7, further comprising repeating said steps (c) and (d) until all said purchasers have been grouped into said first groups or no further purchasers can be selected from said at least one said second group according to predetermined values of said at least one second characteristic.</p>
    <p>9. A method according to claim 7 or 8, wherein purchasers are selected from said at least one said second group according to said predetermined values by starting with either the highest or lowest value of said second characteristic.</p>
    <p>10. A method of purchasing goods and/or services, the method comprising the steps of:-receiving first data collected from a plurality of potential purchasers, said first data representing a plurality of characteristics relating to said purchasers and/or the goods and/or services to be purchased; grouping a plurality of said purchasers into a plurality of first groups by, a) grouping a plurality of said purchasers in a first plurality of second groups according to first predetermined values of at least one first characteristic, and b) selecting purchasers from at least one said second group according to predetermined values of at least one second characteristic to define at least one said first group; and outputting second data in order to seek at least one seller to provide said goods and/or services to at least one said first group.</p>
    <p>11. A method according to claim 10, further comprising:-c) grouping a plurality of said purchasers into a second plurality of second groups according to second predetermined values of at least one said first characteristic, different from at least one said first predetermined values of at least one said first characteristic; and d) selecting purchasers from at least one said second group according to predetermined values of at least one second characteristic to define at least one said first group.</p>
    <p>12. A method according to claim 11, further comprising repeating said steps (c) and (d) until all said purchasers have been grouped into said first groups or no further purchasers can be selected from said at least one said second group according to predetermined values of said at least one second characteristic.</p>
    <p>13. A method according to claim 11 or 12, wherein purchasers are selected from said at least one said second group according to said predetermined values by starting with either the highest or lowest value of said second characteristic.</p>
    <p>14. A method of purchasing goods and/or services substantially as hereinbefore described with reference to the accompanying drawings.</p>
    <p>15. An apparatus for purchasing goods and/or services, the apparatus comprising at least one processor for:-receiving first data collected from a plurality of potential purchasers, said first data representing a plurality of characteristics relating to said purchasers and/or the goods and/or services to be purchased; grouping a plurality of said purchasers into a plurality of first groups according to said characteristics using a solution estimator for an NP-complete problem to determine into which of said first groups a purchaser should be grouped; and outputting second data in order to seek at least one seller to provide said goods and/or services to at least one said first group.</p>
    <p>16. A data structure for use by a computer for purchasing goods and/or services, the data structure comprising:-first computer code executable to receive first data collected from a plurality of potential purchasers, said first data representing a plurality of characteristics relating to said purchasers and/or the goods and/or services to be purchased; second computer code executable to group a plurality of said purchasers into a plurality of first groups according to said characteristics using a solution estimator for an NP-complete problem to determine into which of said first groups a purchaser should be grouped; and third computer code executable to output second data in order to seek at least one seller to provide said goods and/or services to at least one said first group.</p>
    <p>17. A computer running a program according to claim 16.</p>
    <p>18. A data carrier having a program according to claim 16 thereon.</p>
GB0521028A 2005-10-15 2005-10-15 A method of purchasing of goods and/or services Withdrawn GB0521028D0 (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
GB0521028A GB0521028D0 (en) 2005-10-15 2005-10-15 A method of purchasing of goods and/or services

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
GB0521028A GB0521028D0 (en) 2005-10-15 2005-10-15 A method of purchasing of goods and/or services

Publications (2)

Publication Number Publication Date
GB0521028D0 GB0521028D0 (en) 2005-11-23
GB2432232A true true GB2432232A (en) 2007-05-16

Family

ID=35451841

Family Applications (1)

Application Number Title Priority Date Filing Date
GB0521028A Withdrawn GB0521028D0 (en) 2005-10-15 2005-10-15 A method of purchasing of goods and/or services

Country Status (1)

Country Link
GB (1) GB0521028D0 (en)

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2001013300A2 (en) * 1999-08-13 2001-02-22 Demandline.Com, Inc. Aggregation engine
US6356879B2 (en) * 1998-10-09 2002-03-12 International Business Machines Corporation Content based method for product-peer filtering
US20030093355A1 (en) * 1999-08-12 2003-05-15 Gabriel N. Issa, Llc Method, system and computer site for conducting an online auction
US20050209908A1 (en) * 2004-03-17 2005-09-22 Alan Weber Method and computer program for efficiently identifying a group having a desired characteristic

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US6356879B2 (en) * 1998-10-09 2002-03-12 International Business Machines Corporation Content based method for product-peer filtering
US20030093355A1 (en) * 1999-08-12 2003-05-15 Gabriel N. Issa, Llc Method, system and computer site for conducting an online auction
WO2001013300A2 (en) * 1999-08-13 2001-02-22 Demandline.Com, Inc. Aggregation engine
US20050209908A1 (en) * 2004-03-17 2005-09-22 Alan Weber Method and computer program for efficiently identifying a group having a desired characteristic

Also Published As

Publication number Publication date Type
GB0521028D0 (en) 2005-11-23 grant

Similar Documents

Publication Publication Date Title
Rysman The economics of two-sided markets
Besanko et al. Economics of strategy
Liang et al. Importer behavior: The neglected counterpart of international exchange
Bagnoli et al. Durable-goods monopoly with discrete demand
Redding et al. Geography and export performance: external market access and internal supply capacity
Douglas et al. Examining performance of US multinationals in foreign markets
Brousseau The governance of transactions by commercial intermediaries: An analysis of the re-engineering of intermediation by electronic commerce
Gervais et al. The high‐volume return premium
Lounsbury et al. Sources of durability and change in market classifications: A study of the reconstitution of product categories in the American mutual fund industry, 1944–1985
Bhattacharjee et al. Consumer search and retailer strategies in the presence of online music sharing
US20040162757A1 (en) System and method for pay for performance advertising having biddable advertising units utilizing rotating routing to advertiser websites
Montgomery New product distribution: An analysis of supermarket buyer decisions
Nagle Economic foundations for pricing
US20050004835A1 (en) System and method of placing a search listing in at least one search result list
US20060136325A1 (en) Automated proxy bidding
Cosimano Intermediation
US20050228725A1 (en) Methods and systems for selecting and acquiring data to update a geophysical database
Wright et al. Management buy‐outs in the short and long term
Zhang et al. Sticks, carrots, and reforestation investment
Barry et al. Outward FDI and the investment development path of a late-industrializing economy: evidence from Ireland
US20070233561A1 (en) Automated Lead Scoring
US20060069613A1 (en) System for partial automation of content review of network advertisements
Bakker The decline and fall of the European film industry: sunk costs, market size, and market structure, 1890–1927 1
Reardon et al. The supermarket revolution in developing countries: Policies to address emerging tensions among supermarkets, suppliers and traditional retailers
US20070233559A1 (en) Acquiring Leads Using Scoring

Legal Events

Date Code Title Description
WAP Application withdrawn, taken to be withdrawn or refused ** after publication under section 16(1)