Browse Prior Art Database

Auction bid generation using a scheduling system and risk attitudes

IP.com Disclosure Number: IPCOM000018786D
Original Publication Date: 2003-Aug-08
Included in the Prior Art Database: 2003-Aug-08

Publishing Venue

IBM

Abstract

This invention is a methodology and software system for automatically generating recommended actions to supply chain participants in auctions, including forward auctions, reverse auctions, double-sided auctions, combinatorial, and multi-attribute auctions. It is a decision support system that can consider a group of auctions, such as all auctions for products in a given line of business and due in a particular time period. It can operate within the user?s IT environment, thereby revealing no private information to trading partners or intermediaries beyond the bids themselves. The system takes into account existing production conditions such as available production capacity as a function of time, current and predicted production rates and costs, raw material costs as a function of quantity and time, and current inventories of raw materials, work-in-process and finished goods. The system also takes into account current financial conditions, such as the bidder?s cost of money and estimated market prices for its products as a function of time. With these inputs plus the bidder?s estimates of the probability of winning a bid as a function of price, the decision support system generates multiple sets of recommended conditional bids and evaluates them according to bidder-specified criteria. These candidate sets of conditional bids are presented to the bidder for modification and final selection.

This text was extracted from a PDF file.
At least one non-text object (such as an image or picture) has been suppressed.
This is the abbreviated version, containing approximately 16% of the total text.

Page 1 of 10

Auction bid generation using a scheduling system and risk attitudes

Single Parameter Auctions

The figure below depicts the methodology of this invention first for the case of a supply chain bidder participating in an auction in which price is the only negotiable parameter. It is assumed that the bidder is a supply chain manufacturer; i.e. the bidder buys goods and services used in the manufacture of his own end products, which he sells to his customers in a separate process. The case of a supply chain seller is analogous.

Configure Objectives . First the bidder configures the decision support system by defining the business objectives. The bidder might indicate a single objective: maximize profit subject to risk aversion conditions. The system would treat this as a single objective optimization problem, and request the bidder to specify his level of risk aversion (see section "Invoke Bid Generator" below). Alternatively, the bidder might define multiple business objectives, such as profitability, customer satisfaction, and manufacturing disruptions. In either case, the decision support system can make use of the framework described in US Patent 5,940,816, Multi-Objective Decision-Support Methodology.

Select Auctions . The bidder or a filtering mechanism selects a set of auctions in which he might participate. These auctions are for products that the bidder will use to manufacture or deliver his products or services. Thus they could be raw materials, subassemblies, or finished goods that the bidder would resell. They could also include indirect procurement items (maintenance, repair, or operations - MRO) that have an impact on production, such as replacement parts for production equipment. The user can select the auctions by several criteria, including product family, seller, and time period (e.g. auction start time, auction closing time). Such grouping, although optional, would make sense in the case where different product families are manufactured using totally different production equipment and materials, since the bids made for one set of products would be independent of those made for the other.

Get Inventory Positions and Costs . Next the inventory positions and cost data are retrieved for the relevant products. Inventory can include the buyer's own raw materials, work-in-process (WIP), finished goods, spare parts, etc. It can also include data on fixed price inventories owned by other entities in the form of electronic catalogs or the results of a query to an on-line system such as an e-marketplace or private e-commerce Website. Market prices for the buyer's finished products and the cost of money can be used by some algorithms to guide bidding and payment terms.

Get Production Schedules, Rates, and Costs . The current state of the manufacturing plants is extracted from internal systems. This can include open blocks on machines, machine throughput speeds and costs, and variable labor and energy costs. This information can...