Browse Prior Art Database

FORECASTING OR PROCUREMENT OF NEW ASSETS BASED ON ASSET UTILIZATION AND ENTERPRISE COMMERCIAL OR BUSINESS DATA UTILIZING LEARNING ALGORITHMS

IP.com Disclosure Number: IPCOM000237012D
Publication Date: 2014-May-27
Document File: 7 page(s) / 46K

Publishing Venue

The IP.com Prior Art Database

Abstract

The invention proposes a technique to determine number of assets of each type of assets in a hospital. The technique includes a learning algorithm or processes, which includes support vector machine that allows generating decision rules to determine number of assets of each type of assets in the hospital. The algorithm allows procurement based on current level of assets, utilization trend, change in demographics of the region, hospital growth or shrinkage, hospital business plan, evolution of assets and other related factor. The invention proposes a proprietary calculation technique to normalize procurement quantities for evolution in product, availability of competitive product, performance enhancement of future version of same product, change in patient base due to change in region demographics, competition, hospital business plan and other enterprise related factors. The complexity of the proposed algorithm is = O (n2) in the worst case and O (n) in the best case. As a result, solution is scalable. The invention combines heuristic and learning approaches to reliably and accurately predict the asset procurement. There is independence of steps in learning process to enable efficient change or replacement of a sub-process without impacting end outcome.

This text was extracted from a Microsoft Word document.
At least one non-text object (such as an image or picture) has been suppressed.
This is the abbreviated version, containing approximately 25% of the total text.

FORECASTING OR PROCUREMENT OF NEW ASSETS BASED ON ASSET UTILIZATION AND ENTERPRISE COMMERCIAL OR BUSINESS DATA UTILIZING LEARNING ALGORITHMS

FIELD OF INVENTION

The invention generally relates to an asset management in a hospital and more particularly to forecasting and procurement of new assets in the hospital.  

BACKGROUND OF THE INVENTION

An asset in a hospital includes enterprises that consist of patient, physician, nursing, support staff and biomedical equipment, which include intravenous (IV) pumps, ventilators, wheel chairs and patient beds. The present trend in hospital is generalized to any industry, which tends to acquire new or rental assets and is not based on any learning algorithms.  Usually, existing assets do not have any visibilities throughout the enterprise or have minimal visibility in individual department. Any artificial shortage of the assets leads to acquisition of new assets either through purchase, or lease or rental mode of procurement. The procurement is either more or less than requirement and based on experience and is not driven by the data and by right combination of the data. This leads to over or under purchase or rental or lease of the assets. The problem is compounded by a fact that assets acquisition requests are generated from individual department, such as, emergency department or cardiology department and aggregated on the enterprise basis. As a result, any additional or unused assets in one department do not cancel the requirements in another department due to limited wise visibility of the asserts.

Therefore, there is a need for a technique to procure asset driven by data and by correct combination of data.

BRIEF DESCRIPTION OF THE INVENTION

The invention proposes a technique to determine number of assets of each type of assets in a hospital. The technique includes a learning algorithm or processes, which includes support vector machine that allows generating decision rules to determine number of assets of each type of assets in the hospital. The algorithm allows procurement based on current level of assets, utilization trend, change in demographics of the region, hospital growth or shrinkage, hospital business plan, evolution of assets and other related factor.

DETAILED DESCRIPTION OF THE INVENTION

The invention proposes a technique to determine number of assets of each type of assets in a hospital. The technique includes a learning algorithm or processes, which includes support vector machine that allows generating decision rules to determine number of assets of each type of assets in the hospital. The algorithm allows procurement based on current level of assets, utilization trend, change in demographics of the region, hospital growth or shrinkage, hospital business plan, evolution of assets and other related factor. The invention proposes a proprietary calculation technique to normalize procurement quantities for evolution in product, availability of competitive product, performance enhancement o...