DEFINING CYCLE PROTOCOL FOR OPTIMIZATION IN TEXTILE DECONTAMINATION OR CLEANING SYSTEM USING ARTIFICIAL INTELLIGENCE

Description

CROSS-REFERENCE TO RELATED APPLICATIONS

Any new and original work of authorship in this document—including any source code—is subject to copyright protection under the copyright laws of the United States and other countries. Reproduction by anyone of this document as it appears in official governmental records is permitted, but otherwise all other copyright rights whatsoever are reserved.

BACKGROUND OF THE INVENTION

Whilst efficiency improvements have always been of interest to the laundry industry, it is important to note that most of these improvements have been focused on specific equipment units, rather than facility-wide improvements. For example, new washing machines utilize less water and energy. New models of washing machines advertise a level of soiling quantification that may feed information to a laundry dosing system to adjust chemical consumption; however, such quantification is load-specific rather than item-specific and is far less precise than the quantification via imaging stain analysis described here. Moreover, these current systems cannot ascertain the effectiveness of their decontamination cycle. There is no automatic feedback loop. Centralized IT systems employed in laundry facilities are normally focused on reducing downtime, for example by identifying errors in equipment operation or by tracking items in bulk through the decontamination process. The ability to track the entire use and decontamination history of a textile item is a fairly new development, with applications still being developed.

The laundry facilities are seeking to consistently operate more sustainably and efficiently without compromising decontamination results. Improvements in operational efficiency can be obtained by measuring and customizing decontamination protocols to factors that influence textile decontamination. Decontamination textiles is influenced by factors such as inconsistent decision making and human error, soiling intensity, age of soiled items and stains, decontamination equipment in use, water hardness, temperature used in decontamination etc. For every item that is to be cleaned, there is potential to optimize decontamination protocols specific to the textile, level of soiling, facility, decontamination equipment, chemicals and other resources available. Such decontamination protocols would maximize the efficient utilization of resources to reliably clean the item.

One or more embodiments of the invention in accordance with one or more aspects and features are believed to address one or more of such needs.

SUMMARY OF THE INVENTION

Additional aspects and features of the invention relating to validation systems are disclosed in the appendix to the specification attached hereto, which is incorporated herein by reference.

Additional aspects and features are disclosed in U.S. patent application publication 2024/0287723A1, which the disclosure of which is incorporated herein by reference.

In addition to the aforementioned aspects and features of the invention, it should be noted that the invention further encompasses the various logical combinations and subcombinations of such aspects and features. Thus, for example, claims in this or a divisional or continuing patent application or applications may be separately directed to any aspect, feature, or embodiment disclosed herein, or combination thereof, without requiring any other aspect, feature, or embodiment.

BRIEF DESCRIPTION OF THE DRAWINGS

One or more preferred embodiments of the invention now will be described in detail with reference to the accompanying drawings, wherein the same elements are referred to with the same reference numerals.

FIG. 1 is a schematic illustration of a flow of textile items and information in a preferred embodiment in accordance with one or more aspects and features of the invention.

FIG. 2 is a schematic illustration of a learning system in a preferred embodiment in accordance with one or more aspects and features of the invention.

FIG. 2A is a schematic illustration of steps performed in the preferred embodiment of FIG. 2.

FIG. 3 is a schematic illustration of steps of processing textile items for decontamination in a preferred embodiment in accordance with one or more aspects and features of the invention.

FIG. 4 is a schematic illustration of steps of optimizing a decontamination protocol for implementation with textile items in a preferred embodiment in accordance with one or more aspects and features of the invention.

FIG. 5 is a schematic illustration of components of a preferred IT system in a preferred embodiment in accordance with one or more aspects and features of the invention.

FIG. 6 is a schematic illustration of a preferred tracking system in a preferred embodiment in accordance with one or more aspects and features of the invention.

FIG. 7 is a Table “A” of data acquired and logged in a preferred embodiment in accordance with one or more aspects and features of the invention.

FIG. 8 is another Table “B” of data acquired and logged in a preferred embodiment in accordance with one or more aspects and features of the invention.

FIG. 9 is a Table “C” of data types of the data that may be acquired and logged in a preferred embodiment in accordance with one or more aspects and features of the invention.

FIG. 10 is a schematic illustration of the results of multiple performances of the decontamination of a textile item in a preferred embodiment in accordance with one or more aspects and features of the invention.

FIG. 11 is a schematic illustration of steps of the analysis of a decontamination cycle in a preferred embodiment in accordance with one or more aspects and features of the invention.

FIG. 12 is a graph of temperature profile iterations in a preferred embodiment in accordance with one or more aspects and features of the invention.

DETAILED DESCRIPTION

As a preliminary matter, it will readily be understood by one having ordinary skill in the relevant art (“Ordinary Artisan”) that the invention has broad utility and application. Furthermore, any embodiment discussed and identified as being “preferred” is considered to be part of a best mode contemplated for carrying out the invention. Other embodiments also may be discussed for additional illustrative purposes in providing a full and enabling disclosure of the invention. Furthermore, an embodiment of the invention may incorporate only one or a plurality of the aspects of the invention disclosed herein; only one or a plurality of the features disclosed herein; or combination thereof. As such, many embodiments are implicitly disclosed herein and fall within the scope of what is regarded as the invention

Accordingly, while the invention is described herein in detail in relation to one or more embodiments, it is to be understood that this disclosure is illustrative and exemplary of the invention and is made merely for the purposes of providing a full and enabling disclosure of the invention. The detailed disclosure herein of one or more embodiments is not intended, nor is to be construed, to limit the scope of patent protection afforded the invention in any claim of a patent issuing here from, which scope is to be defined by the claims and the equivalents thereof. It is not intended that the scope of patent protection afforded the invention be defined by reading into any claim a limitation found herein that does not explicitly appear in the claim itself.

Thus, for example, any sequence(s) and/or temporal order of steps of various processes or methods that are described herein are illustrative and not restrictive. Accordingly, it should be understood that, although steps of various processes or methods may be shown and described as being in a sequence or temporal order, the steps of any such processes or methods are not limited to being carried out in any particular sequence or order, absent an indication otherwise. Indeed, the steps in such processes or methods generally may be carried out in various different sequences and orders while still falling within the scope of the invention. Accordingly, it is intended that the scope of patent protection afforded the invention be defined by the issued claim(s) rather than the description set forth herein.

Additionally, it is important to note that each term used herein refers to that which the Ordinary Artisan would understand such term to mean based on the contextual use of such term herein. To the extent that the meaning of a term used herein—as understood by the Ordinary Artisan based on the contextual use of such term—differs in any way from any particular dictionary definition of such term, it is intended that the meaning of the term as understood by the Ordinary Artisan should prevail.

With regard solely to construction of any claim with respect to the United States, no claim element is to be interpreted under 35 U.S.C. 112(f) unless the explicit phrase “means for” or “step for” is actually used in such claim element, whereupon this statutory provision is intended to and should apply in the interpretation of such claim element. With regard to any method claim including a condition precedent step, such method requires the condition precedent to be met and the step to be performed at least once but not necessarily every time during performance of the claimed method.

Furthermore, it is important to note that, as used herein, “comprising” is open-ended insofar as that which follows such term is not exclusive. Additionally, “a” and “an” each generally denotes “at least one” but does not exclude a plurality unless the contextual use dictates otherwise. Thus, reference to “a picnic basket having an apple” is the same as “a picnic basket comprising an apple” and “a picnic basket including an apple”, each of which identically describes “a picnic basket having at least one apple” as well as “a picnic basket having apples”; the picnic basket further may contain one or more other items beside an apple. In contrast, reference to “a picnic basket having a single apple” describes “a picnic basket having only one apple”; the picnic basket further may contain one or more other items beside an apple. In contrast, “a picnic basket consisting of an apple” has only a single item contained therein, i.e., one apple; the picnic basket contains no other item.

When used herein to join a list of items, “or” denotes “at least one of the items” but does not exclude a plurality of items of the list. Thus, reference to “a picnic basket having cheese or crackers” describes “a picnic basket having cheese without crackers”, “a picnic basket having crackers without cheese”, and “a picnic basket having both cheese and crackers”; the picnic basket further may contain one or more other items beside cheese and crackers.

When used herein to join a list of items, “and” denotes “all of the items of the list”. Thus, reference to “a picnic basket having cheese and crackers” describes “a picnic basket having cheese, wherein the picnic basket further has crackers”, as well as describes “a picnic basket having crackers, wherein the picnic basket further has cheese”; the picnic basket further may contain one or more other items beside cheese and crackers.

The phrase “at least one” followed by a list of items joined by “and” denotes an item of the list but does not require every item of the list. Thus, “at least one of an apple and an orange” encompasses the following mutually exclusive scenarios: there is an apple but no orange; there is an orange but no apple; and there is both an apple and an orange. In these scenarios if there is an apple, there may be more than one apple, and if there is an orange, there may be more than one orange. Moreover, the phrase “one or more” followed by a list of items joined by “and” is the equivalent of “at least one” followed by the list of items joined by “and”.

Additionally, as used herein, “decontamination” refers to the neutralization or removal of dangerous substances, radioactivity, or germs from an area or object, and may include sterilization or disinfection. Decontamination may be performed by a number of techniques, including heating, application of a sterilizing gas, and radiation. Decontamination also may be performed during and as part of cleaning, including washing.

With respect to aspects and features of the invention relating to “AI”, this term is intended to broadly mean “artificial intelligence” and, in many embodiments, refers to the use of “machine learning”, which refers to programs that can improve their performance on a given task automatically. There are several kinds of machine learning. Unsupervised learning generally refers to analyzing a stream of data and identifying patterns and making predictions without further guidance. “Supervised learning” when used herein generally requires a human to label input data first, and comes in two main varieties: classification, where the program must learn to predict in which category the input belongs; and regression, where the program must deduce a numeric function based on numeric input. “Transfer learning” is considered herein when the knowledge gained from one problem is applied to a new problem; and “deep learning” is considered herein a type of machine learning that runs inputs through biologically inspired artificial neural networks for all of these types of learning. “AI” as used herein thus generally encompasses one or more of these types of machine learning.

Referring now to the drawings, one or more preferred embodiments of the invention are next described. The following description of one or more preferred embodiments is merely exemplary in nature and is in no way intended to limit the invention, its implementations, or uses.

Turning now to FIG. 1, this figure illustrates a flow of textile items and information through a decontamination cycle in a preferred embodiment of a system 101 in accordance with one or more aspects and features of the invention. Additionally, FIG. 2 illustrates a preferred top level learning system 201; FIG. 2A illustrates steps performed in the preferred learning system of FIG. 2; FIG. 3 illustrates a process map from textiles incoming for processing through to assignment to most appropriate machine for processing; and FIG. 4 illustrates a process map for decontamination protocol assignment and recording.

In FIG. 1, solid lines indicate the flow of textile items through a laundry facility, and dashed lines indicate information links and flow through the decontamination cycle.

The system 101 comprises AI and, in particular, an unsupervised or automatic machine learning system for discovering and applying an ideal decontamination protocol for contaminated textile items. The embodiment is tailored to any decontamination equipment and other resources available.

Furthermore, the learning system allows the decontamination protocol to be optimized for the facility operation, through maximizing the balance between decontamination effectiveness, operational economics, and environmental and social impacts. The learning system achieves this by unifying data from three operations of laundry facilities—item tracking, contamination quantification, and process monitoring—through state-of-the-art technology and automation, and by employing a fourth operation to iteratively learn and improve process flow by prescribing bespoke decontamination protocols.

In essence, the learning system, represented as the “Decontamination Cycle Analysis Engine” in FIG. 1, uses a set of cyclical inputs, analyzes them; and outputs an optimized decontamination cycle protocol for next use. With the learning system, optimizations are captured and applied in real-time, with each new decontamination cycle. Therefore, the system is in a state of continuous improvement as part of normal operation.

Furthermore, it will be appreciated that the system may be at any time at any level between maximum improvement and maximum known efficiency. The learning system preferably is an incremental learning system that learns in real time based on data acquired during each cycle decontamination cycle. In contrast, non-incremental learning systems include a costly period of data collection where benefits are delayed. This could be costly, considering the large amount of data needed to train models. Furthermore, non-incremental learning systems are exposed to risks of further data collection and retraining if equipment is changed or operations are significantly altered. The learning system 201 is adaptable to any equipment, set of items, or novel interventions and is able to accelerate improvements by building on decontamination know-how that facility operators have.

A large amount of data inputs preferably is employed to optimize the learning system. In a simplified example, where data is assumed collected and ready for training, then the inputs preferably comprise at least the following inputs, analysis, and outputs.

Inputs

For the Decontamination Cycle Analysis Engine, the inputs preferably include: optimization intentions, e.g., cost reduction, water usage, energy usage, and time savings; tracking and use history of each item; item contamination quantification pre-processing; prescribed decontamination protocol including cycles; actual decontamination protocol log as measured and recorded; item contamination quantification post-wash; history of contamination from previous washes; and a spatial record of the environment and the movement of people, equipment including robotics, textiles and other dynamic features within it.

Analysis

The Decontamination Cycle Analysis Engine within the learning system preferably comprises an elastic weight consolidation (EWC) deep neural network incremental learning system that offers plasticity and does not forget learned benefits. The Image Data Capture preferably incorporates a physical scanning or video system, along with a Convolutional Neural Network (CNN) to export the item stain quantification data listed in the inputs above. The IT System preferably implements predictive analytics with data from Facility and Environmental Data Capture to optimize facility operation, such as scheduling high energy tasks during off-peak energy supply.

Outputs

The outputs from Decontamination Cycle Analysis Engine preferably comprise analytical outputs, automated programming and recommended interventions for optimized:

Categorization/Sorting utilizing textile tracking technology such as RFID, optical tracking and/or machine vision item recognition paired with machine vision to categorize the decontamination by type and quantification (FIG. 3). For example, 5% decontamination of textile composed of content consistent with 1% hemoglobin, 5% fat, 20% protein and 70% water ratios. This categorization could then either inform a conveyor system to place grouped textiles into different machine load tubs for processing or inform an operator which tub they need to place it in or a combination of automation and operator instruction.

Prescribed Optimized Decontamination Protocols for each set of weighed and grouped textiles where the textiles are assigned to the most optimal machine in the facility for decontamination (FIGS. 3 and 4). The textiles could be automatically placed into the machine utilizing robotics or an operator could be informed which machine to place the textiles in. The tagged textiles are then validated that they are in the right machine and then an instruction is sent to the machine processors, which includes the washing machine and dosing pumps, from the cloud processor. Parts of this process could be performed by manual input depending on the level of automation available.

The analysis of use of space and the interactions over time amongst people, equipment, textiles, budgets, cost of goods/building and other dynamic features that would output optimized new layouts, equipment specifications, building and project plans for either optimization of the physical and equipment infrastructure of an existing facility or the creation of new facilities.

Combinations and permutations of the categories, decontamination protocols and space optimization.

New processing center designs utilizing space, budget and operational capabilities.

The decontamination protocol may be prescribed in any format that is applicable to decontamination. It may comprise of an array listing time-series profiles that can guide a user, robot, partially automatic or fully automatic system to program inputs into the decontamination equipment. These may include, for example:

• • Total wash program cycles/phases breakdown including different cycles/phases such as soaking, washing, rinsing, draining and extracting.
  • Chemical dosing profile over time and cycle.
  • Temperature profile over time and cycle.
  • Water inflow and/or outflow profiles over time and cycle.
  • Mechanical wash profile including settings such as mechanical action including rate, frequency and speed of agitation, spinning or combinations of these over time and cycle.

These prescribed protocols are different from the actual protocol logs, which are obtained from the Decontamination Cycle Data Capture, and both preferably are used as feedback inputs in the learning system.

Of course, there are multiple factors that need to be considered when processing textile items for reuse including the type of textile items, the industry it is used in, the regulatory controls within that industry, the level of soiling and the facility, equipment, staff, chemical, water, energy, time and economic resources available. This is of heightened importance within the medical sector, particularly in operating rooms for items such as surgeon gowns and operating theatre drapes which need to be cleaned and sterilized to a level that will minimize the risk of surgical site infection.

When washing and processing items, a balance must be struck between effective decontamination, disinfection, economics, environmental impact and maintaining textile integrity without any compromise on infection control measures or regulatory controls. This balance can vary depending on multiple factors including washing equipment, facility parameters, human factors, textile construction and contamination levels of items. It can also change depending on what are the key performance indicators of the operator, that is, it may be desirable for performance to be optimized for speed of process, cost reduction, waste reduction and so on.

Protocols for wash processes specifying parameters regarding one or more mechanical actions, chemicals, temperature, water level, water quality, and time have been established based on empirical data proving their effectiveness in decontamination and decontamination. A wash process that is effective is considered to be one in which the textiles are sufficiently decontaminated through disinfection defined as removal of infectious agents or colony forming units (CFUs) to an acceptably low amount and decontamination defined as removal of dirt to an acceptably low amount.

The difference between complete decontamination and acceptable decontamination can be utilized to facilitate learning and optimization of processing. There is a demand to improve the efficiency of wash processes such as through reducing the energy requirements for heating water. Lower energy requirements for decontamination can reduce both monetary costs and negative environmental impact through decreased CO2 emissions. Different strategies to reduce energy costs include decreasing the temperature required for decontamination, utilizing heat transfer and scavenging technologies, decreasing water usage and minimizing the mechanical action for effective removal of contaminants. There also has been an increasing demand to use chemicals that have lower negative impacts on both the environment and also the textiles they are decontaminating.

As such, there are multiple factors that can be modified to effectively decontaminate a textile. Furthermore, different textiles, levels and types of contamination all have different requirements for effective decontamination.

Accordingly, in greater regard and detail to preferred embodiment of systems in accordance with one or more aspects and features of the invention, reference is now made to FIGS. 5-11.

In particular, FIG. 5 illustrates components of a preferred information technology (“IT”) system 501 in which APIs (Application Programming Interfaces) and I/O (Input/Output) Systems comprise the interface for receiving and sending data in system 601. The IT system 501 is configured to provide comprehensive data management and visualization capabilities. The IT system 501 includes a software and hardware interface for receiving inputs from various data streams; a central data repository for storage; a processing unit for analyzing and preparing data; decontamination equipment; controllers for the decontamination equipment; and a communication interface for interacting with stakeholders through visual displays, dashboards on web apps, mobile apps, local applications, and combinations thereof. Parts of the input/output hardware may be decentralized to facility sensor units.

To gain further information related to the textile that can be utilized to optimize decontamination, the IT system 501 is configured with appropriate hardware to centralize item tracking information from outside the laundry facility into the IT System 501. Such a system preferably identifies items as they pass through their use cycle and records, for each item, the timestamp of each scan point and scan location. An embodiment of a tracking system 601 relying on RFID scanners across use cycle locations and RFID labels inside items is illustrated in FIG. 6. In particular, FIG. 6 illustrates a preferred tracking system in which arrows indicate the flow of textile items through use and decontamination cycles, and dashed lines indicate information collected at set points and centralized in the preferred IT System of FIG. 5.

An Item Tracking System will be able to generate data entries specific to textile items. The same entries can be obtained from a different tracking system, such as fully passive RFID scanning, or an imaging scanning system with QR codes or bar codes printed on the items, or on their labels. Other embodiments may involve manual inputs of identifiers for each item, with each system method variation introducing a unique level of uncertainty in the data model. For example, the manual scan of items for one step will offer a wider variation in timestamps within each cycle than a bulk scanning method.

FIG. 7 is a Table “A” of data acquired and logged in a preferred embodiment in accordance with one or more aspects and features of the invention and, in particular, a registry scans for item “x”. Specifically, Table A includes a possible embodiment of data obtained from the Item Tracking System of FIG. 6, which preferably is saved on the storage unit of the IT System of FIG. 5. The table is specific to a single item, in this case identified (ID) as “item_x”. The first row presents current status variables that may be of interest to users of the textile item. The registry may contain additional information that is not captured in Table A, such as history of packing of which the item has been a part of, or history of usage and identifiers for said packs and use cases.

Other item information that may be stored includes usage information since last wash and from previous washes, past and current images that can provide stain assessment, staff in contact with the item, decontamination history including decontamination protocols prescribed, actual decontamination protocols followed, failed decontamination cycle attempts, previous decontamination cycle imagery or data logs.

Data recorded in Table A may be relevant for extracting performance features, and therefore may contain inputs for the Decontamination Cycle Data Capture module or Facility and Environment Data Capture from FIG. 1. For example, the times recorded for each scan registry can be used to extract performance features related to the decontamination cycle, such as decontamination cycle duration, time in use duration and so on. In an example of Table A, Time 3 is the time the item is scanned before starting the washing cycle, and Time 4 is the time the item is automatically scanned when taken out of the washing machine equipment. The duration of a washing cycle is approximated from the difference between Time 4 and Time 3. The drying duration is approximated from the difference between Time 5 and Time 4. The difference between Time 5 and Time 3 estimates the decontamination cycle duration of the specific decontamination cycle protocol identified as (CC(n+1) ) in Table A.

Extensive details of decontamination cycles, like (CC(n+1) ), may be stored in a format as illustrated in Table B. In this regard, the table of FIG. 8 includes data acquired and logged in a preferred embodiment in accordance with one or more aspects and features of the invention and, in particular, a registry of decontamination cycles. Such details may include the time record of each decontamination cycle attempt, the equipment used to employ the decontamination cycle, the prescribed sorting of items and the actual items loaded into the cycle. It may also include the prescribed protocol for each decontamination cycle and the actual protocol log collected by the Decontamination Cycle Data Capture module from FIG. 1.

The Decontamination Cycle Data Capture module preferably extracts sensing data that is used to improve the accuracy of duration estimations from Table A, by synchronizing wash cycle duration with energy usage logs for each washing machine (see Table B). Equally this can be done for drying and sterilization. Furthermore, sensors and monitoring methods can be used to measure additional decontamination cycle process monitoring information, as listed in Table C of FIG. 9, which table illustrates data types of the data that may be acquired and logged in a preferred embodiment in accordance with one or more aspects and features of the invention and, in particular, identifies possible data types as wells as methods and apparatus for capturing data with regard to the Decontamination Cycle Data Capture. Details of how the actual protocol log is generated for each type of time series data are listed in Table C.

Within the context of the foregoing disclosure, a preferred method in accordance with one or more aspect and features of the invention comprises:

Step 1

The Facility and Environmental Data Capture feeds up-to-date information to the IT System. The IT system constantly updates information through the Item Tracking System (FIG. 6).

Step 2

Soiled items arrive at the facility at Check-In Items.

Step 3

Used items to be cleaned are extracted for stain evaluation.

Step 4

Imaging 1 Data Capture uses as input one or more textile items and outputs Stain Quantification data (Table 1) into the IT System. This action block may involve capturing a scan of the item(s), analyzing it to offer a stain quantification. The Stain Quantification (Table 1) may involve data comprising of, but not limited to: stain type stain composition, stain size, stain position on the item, link to any previous stains, link to previous item use case, stain age. It may be in the form of a low dimensional vector or other format that could be used as input by the Decontamination Cycle Analysis Engine. Ideally all items to be cleaned are evaluated for stains and the Stain Quantification of all the soiled items is inputted into the IT System. However, the IT system can extrapolate information even if only some or none of the items are being fed through Imaging Data Capture. The IT system may use historical data on item usage (Table 1) and decontamination protocols (Table 2) to estimate soiling levels and generate data to be used as input by the Decontamination Cycle Analysis Engine.

Step 5

The Decontamination Cycle Analysis Engine uses as inputs Stain Quantification of current soiled items to be cleaned and desired optimization (cost, energy efficiency, water efficiency, time efficiency etc.). It outputs a categorization (sorting) into wash loads and decontamination protocols (Table 2 —Prescribed Protocol) for each wash load. The Decontamination Cycle Analysis Engine may give feedback to the IT System on the Prescribed Protocols, the IT system may approve them, or it may resubmit another input. For example, the IT system may change the desired optimization from cost to time efficiency.

Step 6

Sorting and Decontamination Cycle Protocol comprises physically sorting the items into wash loads into the Decontamination Cycle Equipment and setting the prescribed decontamination protocols.

Step 7

Decontamination Cycle Data Capture uses as inputs one or more data streams listed in Table X and outputs a log of the decontamination cycle (Table 2—Actual Protocol Log). This action is to monitor and partly validate the operation of the decontamination cycle.

Step 8

Sorting and Quality Control takes the items out of the decontamination cycle and evaluates their quality against set standards.

Step 9

Imaging 2 Data Capture uses as input one or more textile items and outputs Stain Quantification data (Table 1) into the IT System. It is the same process as before, but this time the items are washed or cleaned. This data is fed back through the IT System to the Decontamination Cycle Analysis Engine.

Step 10

The Decontamination Cycle Analysis Engine uses the newly acquired data to assess the efficacy of its previously prescribed protocols and to apply incremental learning to train itself. The Decontamination Cycle Analysis Engine may give feedback to the IT System on whether decontamination protocols have been improved.

Step 11

The Item Processing takes the clean items and processes them according to set facility protocols.

Step 12

The Check-Out Items block sends out cleaned items to the client.

Facility and environmental information that may be captured by the Facility and Environmental Data Capture may comprise of: staff availability, staff health, staff adherence to QC, quantity of items to be washed, expected time of used item(s) check-in, expected time of clean item(s) check-out, live and predicted energy costs, energy usage, water quality and hardness, clean water usage, wastewater usage, chemicals in stock, cost of chemicals, internal temperatures and humidity, outside temperatures and humidity, local weather data and others.

The Imaging Data Capture module will refer to both elements Imaging Data Capture 1 and Imaging Data Capture 2 from FIG. 3. The module will operate in a similar fashion in both instances. Its role is to capture consistent images that can be linked to items and decontamination cycles. In the presented examples, such images are processed to output stain quantification variables, data that can be stored as in Table 1—Stain Quantification column. Illustrations of this data are presented in FIG. 4.

The Imaging Data Capture will comprise of both hardware and software components, including of:

• • Feed System—the feed system is designed to smoothly transport the item through the scanner. It may include rollers, belts, hangers and rails that guide the item from a pile of clean or soiled items through the scanning system and out of the device without causing damage or alterations to the item. The Feed System must maximize the surface area exposed to the scanning system, for example by rolling the item on both sides, by minimizing creases and opening folds.
  • Optical Scanning System—this is the heart of the scanner, comprising:
  • Light Source—provides consistent uniform illumination necessary to capture the image. LED lights with diffusers are an example.
  • Lens System—employed to focus the reflected or transmissive light with high precision onto the imaging sensor, ensuring sharp and accurate digital representations of the scanned textile item. The quality of the lens affects the sharpness and accuracy of the scanned image.
  • Sensor—typically a Charge-Coupled Device (CCD) or Contact Image Sensor (CIS). The sensor captures the light reflected from the item and converts it into electronic signals. CCD sensors are known for their high quality and color fidelity, while CIS sensors offer a more compact and energy-efficient solution.
  • A simplified optical scanning system comprises ambient room light and one or more cameras capturing photos or videos from a high visibility angle as the items are sorted and quality controlled.
  • Image Processing Hardware/Software—once the image is captured by the sensor, it's processed by the scanning system's onboard hardware and software. This can include adjustments for color, contrast, resolution, and correction of any distortions or artefacts introduced during the scanning process. Advanced scanning systems might also have features for automatic item feature recognition, image straightening (deskew), stain identification and quantification.
  • Output Interface—this component manages the transfer of the scanned image data from the scanning system to a computer or storage device before ultimately reaching the dashboard. It can include connections like Bluetooth, USB, Ethernet, or Wi-Fi.

FIG. 10 is a schematic illustration of the results of multiple performances of the decontamination of a textile item in a preferred embodiment in accordance with one or more aspects and features of the invention, wherein stain quantification of a single item is shown before and following each of a different number of decontamination cycles. In FIG. 10, a textile item or items is/are presented with a stain, S1, which is quantified in shape, in location on the item, and intensity—illustrated as opacity in this example. After decontamination cycle CC(1) is performed, the stain S1 is 95% clean, illustrated by the 5% opacity. Assuming the threshold for a quality control pass is 90%, then the stain decontamination assessment is considered passed. This is not ideal, as S1 continues to be visible on the item throughout its lifecycle. But the Imaging Data Capture records the stain information and attributes it to the original first use, and first wash.

In contrast, stain S2, from the second use cycle is not cleaned to quality control standards after the decontamination cycle CC(2). It is assumed it is cleaned above 90% in the subsequent cleans after CC(2). Stain S3 is completely cleaned by CC(2). This type of Stain Quantification information will be used as input for the Decontamination Cycle Analysis Engine module. In this embodiment, the purpose of the module will be to learn from such stain quantification information in order to prescribe future decontamination protocols that will improve stain removal efficiency.

An embodiment of the Decontamination Cycle Analysis Engine, FIG. 11 comprises multiple individual entities that will work in-sync to produce the best overall washing parameters. In this respect, FIG. 11 is a schematic illustration of steps of the analysis of a decontamination cycle in a preferred embodiment in accordance with one or more aspects and features of the invention and, in particular, illustrates components of the Decontamination Cycle Analysis Engine, with stain quantification data extracted from within the Image Data Capture (outlined block at the top). The parameters not only are better suited for decontamination of the textile but also reducing the overall cost of the operation. The cost can be composed of features like when to wash the textiles, effective load for decontamination the textiles etc.

The first part of this engine will comprise of a “Stain Classification System” that will categorize the stain and produce the localized stain information (where the stain is on the fabric and the intensity of the stain) in the form of a vector. “Stain Classification System” takes a combination of Image data and sequential data as input to categorize the stain. This vector will then be used by a “Parameters Predictor” to generate wash cycle parameters (Prescribed Decontamination Cycle Protocol) like spin speed, temperature, chemical composition. These parameters are learned using the previous washing history as per the standards and full load capacity of the machine. The images of textiles taken at multiple intervals can be an example of the sequential data. The “Parameters Predictor” will also receive composition information along with the stain information from “Stain Classification System” which is utilized to obtain best possible parameters given the composition of the fabric.

The “Cost Optimizer” stage takes the generated wash parameters as input along with the absolute information like cost of electricity, cost of water, environmental impact of chemicals etc. to produce a vector of new information that minimizes the overall set cost of the operation. This new vector is analyzed with the “Parameters Predictor”output and the saving information is logged for future use.

This is the last operation of the system and the result of this is a best vector of Prescribed Decontamination Cycle Protocol (wash cycle parameters). This is used by the washer machines to wash the textiles. After the decontamination process the washed textiles are then analyzed by the Image Data Capture 2 (FIG. 3), which is periodically validated by a human which either marks the clean cycle as “Pass or Fail”. In case of failure the whole process from either the “Parameters Predictor” is repeated to tweak the parameters and generate the new best vector for wash, or the process is repeated from the “Stain Classification System” to recategorize the stain. This process of selecting the appropriate point of re-training (Parameters Predictor, Stain Classification System) is selected by a human response.

Each step's output and input are logged for future analysis study. These stored records are also used by the dashboard application to allow users to analyze savings and history of the washes over a period.

For every decontamination equipment, load combination and time of operation there are a set of theoretical ideal decontamination cycle protocols. One ideal protocol for minimal energy use, one of minimal water use, and one of minimal chemical use and so on. There is also a theoretical ideal decontamination cycle that balances consumption of all employed resources with respect to cost considerations and/or environmental harm minimization.

The Learning System is designed to identify these theoretical ideal decontamination cycle protocols, to identify practical ideal decontamination cycle protocols (with safety margins within which to perform decontamination reliably) and to prescribe parameters of operation to lower cost, lower environmental impact and increased efficiency of decontamination operations.

These ideal theoretical and practical decontamination cycle protocols could be computed from infinite data specific to each facility, equipment, textile item, if accurately collected and made available. In practice, such data is always limited by unavailability and inaccuracies in collection methods.

Preferred embodiments of the invention provide methods to increase availability and accuracy of relevant data to textile decontamination operations. More importantly, the disclosure provides means to employ limited available data to enact operational efficiency benefits geared towards cost reduction, staff welfare, environmental harm reduction. Attempting to push cost/energy/water/chemical towards the consumption of the ideal theoretical decontamination cycle will result in instances of improper, ineffective decontamination. Therefore, the Learning System will operate somewhere between the following two modes:

• • Learning operation—where boundaries of consumption limits are pushed to edge closer to the ideal protocol. This involves more risk and higher costs for mitigation; and
  • Efficient operation—where it operates within a safety margin to provide reliable decontamination protocols. The safety margin should compensate for any variations in operational parameters of equipment and/or environment. The Learning System will use previous results, data and user-imposed constraints to define operation.

Example of Learning Operation for One Parameter (Temperature)

The last decontamination cycle protocol known to be effective, from previous operations or from user input, will be named CC0(T). (T) is the temperature profile of the decontamination cycle, which is represented by the solid green line in FIG. 12. In this regard, FIG. 12 is a graph of temperature profile iterations in a preferred embodiment in accordance with one or more aspects and features of the invention, and in particular, illustrates examples of temperature profile iterations throughout the learning operation. In FIG. 12, the assumption is made that the resolution of the equipment, i.e., increment, is 3 degrees Celsius.

A user-imposed constraint requires the decontamination cycle to incorporate a wash stage at 71 C that is maintained for a minimum of three minutes. This remains in effect during the entirety of the learning operation. The Learning System will attempt to lower the T (Temperature) by increments with each new decontamination cycle. The first decontamination protocol suggested will be CC0(T−1), where T−1 is the temperature profile decreased by one increment; the light blue solid line in FIG. 12. Stain comparison from visual data captured pre and post wash will reveal if CC0(T−1) is effective in washing soiling items. If it is effective, CC0(T−2) will be trialled next, where temperature is decreased by two increments, and so on. If CC0(T−2) is deemed ineffective, the Learning System will propose CC1(T) where temperature, i.e. consumption, increases compared to CC0(T−2) , but remains smaller than CC0(T−1).

For a specific case example, consistent with FIG. 12, the following learning steps are presented:

• • CC0(T), pulled from historical operations, involved a wash load of 40 items with 110 registered stains pre-wash, 100% effectiveness in stain removals with 0 fails. Cycle protocol CC0(T) is used to generate CC0(T−1) CC0(T−1) is employed with 40 items in the wash load that registered 102 stains. 100 stains are removed completely and 2% of the recorded 2 stains remain, then those stains are recorded as fails.
  • Cycle protocol CC0(T−1) is deemed effective, as the persistent stains are >90% washed.
  • CC0(T−2) is employed with 40 items in the wash load that registered 99 stains. 80 stains are removed completely and the 19 fails on 9 items have some stains that are less than 90% washed.
  • Cycle protocol CC0(T−2) is deemed ineffective. CC0(T−3) will not be trialed. CC1(T) is prescribed instead.
  • CC1(T) is employed with 40 items in the wash load that registered 100 stains. 99 stains are removed completely. 1 stain remains persistent, but over 90% washed.
  • Cycle protocol CC1(T) is deemed effective.
  • CC1(T) is now adopted as the most efficient decontamination protocol to be employed in Efficient operation.

This example employs some simplifications for ease of understanding. It presents wash loads that are very similar and comparable, consisting of 40 items of the same lifecycle stage (e.g., 20th wash cycle out of 75). During operation, items at an earlier lifecycle stage may require different decontamination cycle protocols than items at the later lifecycle stages. The Learning System will account for that. The Learning System will also account for the estimated age of each stain.

The example only iterates on the temperature profile. In effect, the Learning System may choose to iterate on multiple parameters at once, including, but not limited to spin cycle, chemical dosage, water quantity, drying protocol and others.

The example presents an incremental learning model, where data is being iteratively captured in the Learning mode, on a prescribed decontamination cycle. In effect, batch learning models can also be employed. Data will be captured during operational modes as well, and on failed/incomplete cycles. For example, CC0(T) may be started but may fail completion at minute 80 instead of minute 90. The items may continue to the Imaging Data Capture for stain examination. If the stain data evaluation reveals high effectiveness of CC0(T) cut at minute 80 instead of minute 90, then the Learning System may adopt CC0(T) cut at minute 80 as a new effective decontamination protocol.

The example also presents stain information being captured before and after decontamination. It may be that no stain information is captured pre-wash, but the Learning System will still estimate efficiency of decontamination protocols, albeit with greater uncertainty. Visual data capture and Stain information may be limited to a selection of items going through the decontamination cycle, or to none at all. The Learning System can employ historical data on decontamination cycle effectiveness, and predicted soiling information based on information such as, but not limited to where the items were used, how long ago they were used, how many wash cycles have they been through, the protocols employed in the previous wash cycles, what material they are made of, which supplier provided them, and others.

The example presents the last effective decontamination protocol as the default decontamination protocol during operation. During operation, there may be a host of protocols that the Learning System may choose from. For example, a shorter duration decontamination cycle may be chosen, despite higher energy use, if the Learning System anticipates high energy operations may overlap with live energy cost increases, or if it anticipates staff working hours will finish before decontamination cycle operation.

Example of Efficient Operation

One hundred and sixty items are checked in for decontamination. The tracking information, centralized in the Dashboard, indicates they come from 8 surgeries, which have occurred in the past week, all within the same hospital. The earliest time of soiling (surgery) was 6 days ago, the latest 1 day ago.

Before Sorting, the Learning System suggests four Decontamination Cycle Protocols using information from the life cycle of checked in items, as well as previous soiling imaging information from past washes and the estimated level of soiling expected from the surgeries the items were employed in.

The Learning System suggested 4 Decontamination Cycle Protocols are:

• • For Washing Machine 1: CC1.0 (40 items), start at 10 AM, 1 h 30 m duration, Chemical dosage (50%) then CC3.0 (57 items), start at 11:45 AM, 1 h 40 m duration, Chemical dosage (75%).
  • For Washing Machine 2: CC2.0 (38 items), start at 10 AM, 1 h 30 m duration, Chemical dosage (50%), then CC4.0 (25 items), start at 11:45 AM, 2 h 15 m duration, Chemical dosage (80%).

During Sorting, visual images of soiled items are captured and soiling information is quantified. It emerges that fewer stains than expected by the Learning System were produced in this use cycle.

The Learning System updates the CC Protocol prescriptions to improve energy efficiency and reduce cost:

• • For Washing Machine 1: CC1.1 (50 items), start at 10 AM, 2 h 00 m duration, Chemical dosage (85%), then CC3.1 (58 items), start at 12:25 PM, 2 h 25 m duration, Chemical dosage (95%)
  • For Washing Machine 2: CC2.1 (52 items), start at 10 AM, 2 h 00 m duration, Chemical dosage (85%)

In this case, the Learning System saved energy usage by avoiding one decontamination cycle operation (CC4.0) but using more chemical compounds in the remaining 3 decontamination cycles. The Learning System also saved costs by recommending the high energy decontamination cycle (CC3.1) start at a later off-peak time, when energy costs are lower.

Stain examination is only performed on CC3.1. It reveals that efficiency was below expected ranges, items with stains older than 5 days were not cleaned effectively. Data is being stored to be used in the next batch Learning System update.

From the foregoing, it will now be appreciated that preferred embodiments of the invention offer uniquely comprehensive systems for continuously improving wash cycle and decontamination efficiency via textile soiling quantification and wash cycle monitoring. Rather than focusing on stain removal, preferred embodiments of the invention focus on process optimization utilizing many different feedback data streams.

Based on the foregoing description, it will be readily understood by those persons skilled in the art that the invention has broad utility and application. Many embodiments and adaptations of the invention other than those specifically described herein, as well as many variations, modifications, and equivalent arrangements, will be apparent from or reasonably suggested by the invention and the foregoing descriptions thereof, without departing from the substance or scope of the invention. Accordingly, while the invention has been described herein in detail in relation to one or more preferred embodiments, it is to be understood that this disclosure is only illustrative and exemplary of the invention and is made merely for the purpose of providing a full and enabling disclosure of the invention. The foregoing disclosure is not intended to be construed to limit the invention or otherwise exclude any such other embodiments, adaptations, variations, modifications or equivalent arrangements, the invention being limited only by the claims appended hereto and the equivalents thereof.