Browse Prior Art Database

Cognitive Computing Based Smart Design Option Tool

IP.com Disclosure Number: IPCOM000243454D
Publication Date: 2015-Sep-22
Document File: 3 page(s) / 53K

Publishing Venue

The IP.com Prior Art Database

Abstract

Described is an application of natural language processing to design engineering and brainstorming.

This text was extracted from a PDF file.
This is the abbreviated version, containing approximately 51% of the total text.

Page 01 of 3

Cognitive Computing Based Smart Design Option Tool

The system described in this article is the application of cognitive computing, natural language processing, machine learning, and statistical analysis to create a tool giving design options to engineers based upon analogy to other industries and patented work. The cognitive computing design option tool shifts the task of brainstorming to the computer, allowing the designer to focus on implementing the chosen option.

    The current role of a design engineer is to first learn about the problem trying to be solved, brainstorm options of how the issue can be resolved, assess the different options that arise during the brainstorm, and finally implement the best solution. Often times, brainstorming is a long drawn out process which is not effective in finding all possible solutions to the problem.

Not finding all of the possible solutions, in some cases, leads to a less than optimal solution being implemented to solve a problem.

    Additionally, most, if not all, of the solutions brainstormed by the design team would not be novel mechanisms; they would be mechanisms found in other products which meet the criteria presented. Until now, the best tool for finding those mechanisms from other applications and understanding if they would work well in the desired application is the human brain. Cognitive computing with natural language processing, machine learning, and statistical analysis allows the computer to understand the presented design problem, find products in other applications meeting the defined criteria, and report back to the design team with both an array of all the relevant design options along with a calculated order of the best solutions as defined by relevance to the input criteria.

    This invention consists of the use of a cognitive computer with the ability to process and comprehend natural language to search for relevant solutions to design problems. The computer would ingest a written description of the design problem and use natural language processing to determine the goals and constraints of the problem. The computer would then sort through a database of predefined solutions in other applications derived from freely available, open data (internet, Wikipedia*, patent documents, etc.) using the same natural language processing. Using statistical analysis, the computer would be able to calculate the percent relevance of all possible options and report to the design team which solutions are the best for their application as well as the source material the decision was based on. The design team is thus provided a very complete list of the possible solutions to their design problems as well as evidence-based rankings in a much shorter time frame by leveraging cognitiv...