Method and system for providing an inference engine including a parameter-based cost function to evaluate semiconductor clients
AssigneeTaiwan Semiconductor Manufacturing Company, Ltd.
InventorsYu Jen Chen, Jiann-Yeh Ou
Life of this patent
Estimating the term…
Abstract
An inference engine is configured to rank a plurality of clients using at least one parameter associated with each of the plurality of clients. The engine includes a user interface configured to enable a user to select the at least one parameter, a knowledge collection module configured to collect client information based on the at least one parameter, and a calculation module configured to receive the collected information and calculate a client listing using a parameter-based cost function.
Claims22
A computer program product that includes a computer-readable medium, the medium having stored thereon a sequence of instructions which, when executed by a processor, causes the processor to execute a method, the computer program product comprising:
- an inference engine communicably coupled to a virtual fab, wherein the virtual fab comprises a plurality of entities that are communicably coupled via a network, and a database that stores client information, and the virtual fab is configured to:
- receive integrated circuit manufacturing orders from a plurality of clients;
- and
- facilitate access to client information by the plurality of clients and the entities as the entities produce the integrated circuit manufacturing orders;
- and
- wherein the inference engine is configured to rank the plurality of clients using at least one parameter associated with each of the plurality of clients, the inference engine comprising:
- a knowledge collection module configured to collect client information based on the at least one parameter;
- and
- a calculation module configured to receive the collected information and calculate a client listing ranking the plurality of clients using a parameter-based cost function;
- and
- a user interface operable to:
- enable a user to select the at least one parameter associated with each of the plurality of clients;
- enable the user to assign at least one weighting factor to the selected at least one parameter associated with each of the plurality of clients;
- display the client listing calculated by the calculation module,
- wherein the calculation module calculates the client listing based on the selected at least one parameter and the assigned at least one weighting factor for the at least one parameter,
- wherein the parameter-based cost function comprises the at least one parameter,
- wherein the parameter-based cost function is a linear function including at least one term, wherein each term is a product of one of the selected at least one parameter and one of the assigned at least one weighting factor correspondingly, and
- wherein the assigned at least one weighting factor is associated with a weighting category selected from the group consisting of process technology, technology file download, library download, chip implementation, reference flow download, and new tapeout.
A computer program product that includes a computer-readable medium, the medium having stored thereon a sequence of instructions which, when executed by a processor, causes the processor to execute a method of calculating a client listing using an inference engine communicably coupled to a virtual fab, wherein the virtual fab comprises a plurality of entities that are communicably coupled via a network, and a database that stores client information, and the virtual fab is configured to:
- receive integrated circuit manufacturing orders from a plurality of clients;
- and
- facilitate access to client information by the plurality of clients and the entities as the entities produce the integrated circuit manufacturing orders;
- and, the method comprising:
- selecting parameters associated with each of the plurality of clients;
- assigning weighting factors to the selected parameters;
- collecting knowledge relating to the selected parameters;
- selecting a time scope and a geographical region scope, wherein the time scope represents a duration of time for collecting knowledge related to the selected parameters associated with each of the plurality of clients, wherein the geographical region scope represents a geographic region designating a location of at least one client from the plurality of clients;
- and
- calculating a parameter-based cost function from the collected knowledge;
- and
- providing a client listing, wherein clients are ranked according to values of the parameter-based cost function, the selected time scope, and the selected geographical region scope,
- wherein the parameter-based cost function comprises the selected parameters, and
- wherein the parameter-based cost function is a linear function including a term, wherein each term is a product of one of the selected parameters and one of the assigned weighting factors, each of the assigned weighting factors associated with a weighting category selected from the group consisting of process technology, technology file download, library download, chip implementation, reference flow download, and new tapeout.
A system comprising:
- a virtual fab that manufactures one or more products for a plurality of clients, wherein the virtual fab comprises a plurality of entities that are communicably coupled via a network, and a database that stores client information, and the virtual fab is configured to:
- receive integrated circuit manufacturing orders from a plurality of clients;
- and
- facilitate access to client information by the plurality of clients and the entities as the entities produce the integrated circuit manufacturing orders;
- and
- wherein at least one parameter is associated with each of the plurality of clients;
- and
- a computer program product that includes a computer-readable medium, the medium having stored thereon a sequence of instructions which, when executed by a processor, causes the processor to execute a method, the computer program product comprising:
- an inference engine communicably coupled to the virtual fab, wherein the inference engine is configured to rank the plurality of clients using the at least one parameter, the inference engine comprising:
- a knowledge collection module configured to collect client information based on the at least one parameter;
- a calculation module configured to receive the collected information and calculate a client listing ranking the plurality of clients using a parameter-based cost function;
- and
- a user interface operable to:
- enable a user to select the at least one parameter associated with each of the plurality of clients;
- enable the user to assign at least one weighting factor to the selected at least one parameter associated with each of the plurality of clients;
- display the client listing calculated by the calculation module,
- wherein the calculation module calculates the client listing based on the selected at least one parameter and the assigned at least one weighting factor for the at least one parameter,
- wherein the parameter-based cost function comprises the at least one parameter,
- wherein the parameter-based cost function is a linear function including at least one term, wherein each term is a product of one of the selected at least one parameter and one of the assigned at least one weighting factor correspondingly, and
- wherein the assigned at least one weighting factor is associated with a weighting category selected from the group consisting of process technology, technology file download, library download, chip implementation, reference flow download, and new tapeout.
Dependencies are inferred from the claim text.