Eco-efficiency (sustainability) dashboard for semiconductor manufacturing

Abstract

To provide a method and a system for monitoring, correcting, and optimizing the eco-efficiency of semiconductor manufacturing processes and facilities.SOLUTION: A method includes processing logic receiving a first selection of at least one of a first manufacturing process or first manufacturing equipment for performing manufacturing operations of the first manufacturing process, inputting the first selection into a digital replica of the first manufacturing equipment, the digital replica determining environmental resource usage data indicative of a first environmental resource consumption of the first manufacturing process performed on the first manufacturing equipment based on physical conditions of the first manufacturing process output by the first manufacturing equipment, a processing device determining modifications to the first manufacturing process that reduce the environmental resource consumption of the first manufacturing process performed on the first manufacturing equipment, applying the modifications to the first manufacturing process, and providing the modifications for display by GUI.SELECTED DRAWING: Figure 6

Description

[Technical Field]

[0001] This specification relates generally to the environmental impact of semiconductor manufacturing facilities. More particularly, this specification relates to monitoring the ecological efficiency / eco-efficiency of, identifying modifications to, and implementing optimization of semiconductor manufacturing processes and semiconductor manufacturing facilities that perform functions associated with the manufacturing processes. [Background technology]

[0002] The continuing demand for electronic devices requires an ever-greater demand for semiconductor wafers. The increased manufacturing required to produce these wafers places a significant burden on the environment in the form of resource utilization and the creation of environmentally damaging waste. Therefore, there is an increasing demand for greener and more environmentally responsible methods of wafer manufacturing, and manufacturing in general. Given that wafer processing is energy intensive, there is value in decoupling the growth of the semiconductor industry from its environmental impact. Summary of the Invention

[0003] The following is a simplified summary of the present disclosure to provide a basic understanding of some aspects of the disclosure. This summary is not an extensive overview of the disclosure. It is not intended to delineate the full scope of particular embodiments of the disclosure or the full scope of the claims. Its sole purpose is to present some concepts of the disclosure in a simplified form as a prelude to the more detailed description that is presented later.

[0004] Methods and systems, and associated operations of the methods and systems, are described for identifying modifications to a semiconductor manufacturing process associated with changing (e.g., improving) an eco-characteristic (e.g., reducing per-unit environmental resource consumption) of a manufacturing process performed by a semiconductor manufacturing facility. In some embodiments, the method may include receiving, by a processing device, a first selection of at least one of a first manufacturing process or a first manufacturing facility for performing manufacturing operations of the first manufacturing process. The method may further include inputting the first selection into a digital replica of the first manufacturing facility, the digital replica outputting physical conditions of the first manufacturing process. The method may further include determining environmental resource usage data indicative of a first environmental resource consumption (e.g., per-unit environmental resource consumption) and / or an environmental impact (e.g., gas or particulate species entering the atmosphere) of the first manufacturing process performed on the first manufacturing facility based on the physical conditions of the first manufacturing process. The method may further include determining a modification to the first manufacturing process that reduces per-unit environmental resource costs of the first manufacturing process performed on the first manufacturing facility. The method may further include at least one of applying the modification to the first manufacturing process or providing the modification for display by a graphical user interface (GUI).

[0005] In some embodiments, a method is implemented for training a machine learning model to identify modifications to a manufacturing process or a selection of manufacturing equipment for performing manufacturing operations of the manufacturing process. The method includes generating training data for the machine learning model. Generating the training data may include identifying first training inputs having a first selection of a first manufacturing process and identifying a first target output for the first training inputs. The first target output includes a first modification to the first manufacturing process that, when applied, reduces environmental resource consumption and / or environmental impact per first unit of the first manufacturing process. The method further includes providing, by a computing device, the training data to train the machine learning model on a set of training inputs including the first training inputs and a set of target outputs including the first target output. The trained machine learning model may later receive as input a new selection of a new manufacturing process and produce new outputs based on the new inputs, the new outputs indicating new modifications to the new manufacturing process that, when applied, reduces the new environmental resource consumption and / or environmental impact of the new manufacturing process.

[0006] The aspects and embodiments of the present disclosure will become more fully understood from the detailed description provided below and the accompanying drawings, which are intended to illustrate the aspects and embodiments by way of example and not by way of limitation. [Brief explanation of the drawings]

[0007] [Figure 1] FIG. 1 is a block diagram illustrating an example system architecture in which embodiments of the present disclosure may operate. [Figure 2] FIG. 1 is a block diagram illustrating an eco-efficiency sustainability system architecture in which embodiments of the present disclosure may operate. [Figure 3] FIG. 1 is a flow diagram of an exemplary methodology for monitoring, sustaining, and / or optimizing a manufacturing process. [Figure 4] FIG. 1 depicts an exemplary digital replica, according to some embodiments of the present disclosure. [Figure 5] FIG. 1 illustrates an example illustration of a process parameter value window, according to some embodiments of the present disclosure. [Figure 6] FIG. 1 illustrates an exemplary illustration of a method for identifying modifications to a manufacturing process, according to some embodiments of the present disclosure. [Figure 7A] FIG. 1 is a flow diagram of a method associated with identifying modifications to a manufacturing process, according to some embodiments of the present disclosure. [Figure 7B] FIG. 1 is a flow diagram of a method associated with identifying modifications to a manufacturing process, according to some embodiments of the present disclosure. [Figure 7C] FIG. 1 is a flow diagram of a method associated with identifying modifications to a manufacturing process, according to some embodiments of the present disclosure. [Figure 8] 1 is a block diagram of an example computing device that operates in accordance with one or more aspects of the present disclosure. DETAILED DESCRIPTION OF THE INVENTION

[0008] Ecological efficiency (eco-efficiency) characterization is a complex technique used to determine how different levels of inputs (e.g., resources, utilization, etc.) associated with a particular manufacturing tool during its use impact the tool's eco-efficiency. Eco-efficiency characterization can be useful during manufacturing tool development to help develop manufacturing tools that maximize per-unit (or per-hour) eco-efficiency and minimize adverse environmental impacts. Eco-efficiency characterization can also be useful after tool development, when the tool is operable to fine-tune the tool's per-unit eco-efficiency characteristics by taking into account specific parameters according to which tool it is operating.

[0009] Embodiments described herein provide a system for systematically performing eco-efficiency characterization of a manufacturing tool throughout its design, development, and manufacturing process. The embodiments further provide for the integration of eco-efficiency characterization and optimization of multiple processes (e.g., semiconductor devices (e.g., memory, logic, cumulative consumption per integrated circuit (IC)). The embodiments further provide for using eco-efficiency characterization using digital replicas associated with manufacturing processes and / or manufacturing equipment to further determine modifications and / or optimizations to reduce environmental resource consumption and / or environmental impact (e.g., per device, die, wafer, etc.).

[0010] In some embodiments, eco-efficiency is calculated per unit. Typically, per-unit eco-efficiency is not considered in the manufacturing tool development process. Additionally, characterizing per-unit eco-efficiency to adjust settings for a manufacturing tool while the tool is in use (e.g., while the tool is being used for wafer production) can be a tedious and complicated process. Furthermore, previous solutions have used specialized human eco-efficiency training and specialized engineers and analysts for eco-efficiency characterization analysis. Embodiments of the present disclosure provide improved methods, systems, and software for per-unit eco-efficiency characterization. These methods, systems, and software can be used by individuals without specialized eco-efficiency training.

[0011] In one embodiment, eco-efficiency characterization can be performed by a software tool at all stages of the fabrication equipment life cycle, including during the design and operation stages of the wafer fabrication equipment. Eco-efficiency can include the amount of environmental resources (e.g., electrical energy, water, gas, etc.) consumed per unit of equipment production (e.g., per wafer or per device manufactured). Eco-efficiency can also be characterized as the amount of environmental impact (e.g., CO2 emissions, heavy metal waste, etc.) generated per unit of equipment production.

[0012] A per-unit analysis, where the unit is any measurable quantity (e.g., substrate (wafer), die, area (cm), time period, device, etc.) operated by a manufacturing tool, allows for a more precise characterization of eco-efficiency. Eco-efficiency "per unit" allows for a precise determination of resource usage and environmental impact per unit produced and can be easily manipulated as a measure of value. For example, a particular manufacturing tool may be determined to have an eco-efficiency rating per wafer pass of 1.0-2.0 kWh of electrical energy per wafer pass (in other embodiments, the eco-efficiency rating may be less than 0.5 kWh, up to 20 kWh, or even greater than 20 kWh per wafer pass), indicating that each wafer operated by the manufacturing tool may use, for example, 1.0-2.0 kWh of electrical energy per wafer pass. In other embodiments, various other amounts of electrical energy may be used. Determining eco-efficiency per wafer pass allows for easy comparison with other manufacturing tools that have different annual electrical energy consumption values ​​due to variations in annual wafer throughput. In one embodiment, eco-efficiency can also be determined per device by dividing the eco-efficiency characterization per wafer by the number of devices per wafer.

[0013] Conducting eco-efficiency characterization during the early design stages of equipment manufacturing allows designers to make better and more eco-efficient design choices at minimal cost. Eco-efficiency can be manipulated and improved early in the design stage of manufacturing equipment. Early eco-efficiency characterization in the design process can enable better and more eco-friendly component selection, subsystem design, system integration, process design, process material selection, and system configuration.

[0014] In one embodiment, multiple designers may have parallel access to a database of already calculated eco-efficiency models for a particular facility or sub-component. The designers may select one or more sub-components, each of which may have its own eco-efficiency model, and add them together to produce a predicted design. All of the combined eco-efficiency models of the sub-components may then be combined to produce an overall eco-efficiency model for the predicted design. The predicted design and its eco-efficiency model, as well as the eco-efficiency models of its sub-components, may be stored in a database.

[0015] At any time in the tool development process, an engineer can change the tool's configuration, which can cause a change in the eco-efficiency model for the tool. Changes to the configuration and the resulting changes to the eco-efficiency model can be stored in a database. In this manner, eco-efficiency characterization (e.g., per-unit eco-efficiency) can be collaborative, allowing equipment designers to benefit from each other's work. In one embodiment, designers can see updates to the manufacturing equipment design in real time as changes associated with eco-efficiency are made. Designers can select equipment or sub-components with the desired eco-efficiency for a desired application. Furthermore, eco-efficiency (e.g., per-unit eco-efficiency) can be calculated for the manufacturing equipment based on known eco-efficiency characterizations (e.g., per-unit eco-efficiency characterizations) for the sub-components. Such known eco-efficiency characterizations for the sub-components can be stored in a database. In another embodiment, eco-efficiency can be calculated for the manufacturing equipment's combined utility and utilization data for each of the manufacturing equipment's sub-components.

[0016] Components and sub-components can be compared and contrasted. If an eco-efficiency model does not already exist for a particular piece of equipment or sub-component, the designer can perform an eco-efficiency analysis on the equipment and store the resulting eco-efficiency model in a database. Designers may have the option to save various versions of equipment under development, with each version having an associated eco-efficiency model. In this manner, versioning is trackable, and eco-efficiency can be optimized by determining the equipment design version with the desired eco-efficiency. In some embodiments, comparisons between components and sub-components can be used to determine patterns, issues, and / or findings regarding chamber-to-chamber matching between multiple devices with multiple chamber matching. As a result of such comparisons and contrasts between components and sub-components or versions, overall eco-efficiency performance, consumption savings such as carbon footprint, etc. can be reported.

[0017] Manufacturing equipment and subsystems are sometimes used in a variety of applications, each with its own eco-efficiency. In such situations, multiple eco-efficiency characterizations for the same equipment or subcomponent used under different conditions may be stored in the database. When a designer selects an appropriate equipment from the database, various applications for the equipment, each with its own eco-efficiency characterization, may be presented. Furthermore, a designer may select equipment from the database to use as a starting point for a new application that does not yet exist in the database. The designer may modify the equipment's parameters to fit the appropriate application, perform an eco-efficiency characterization, and store the results back in the database.

[0018] In another embodiment, the eco-efficiency characterization may be performed on the manufacturing facility itself during operation. The manufacturing facility may have access to real-time variables, such as equipment utilization and utility usage data, and use the real-time variables in the eco-efficiency model. In this embodiment, the manufacturing facility may fine-tune settings on the facility to maximize eco-efficiency given the current operating conditions of the manufacturing facility. The eco-efficiency characterization on the facility may be useful for fine-tuning the eco-efficiency of a manufacturing facility designed using theoretical, average, or expected variable conditions.

[0019] In some embodiments, eco-efficiency characterization can be determined using a digital replica. Selections, including one of a manufacturing process or a manufacturing facility for performing a manufacturing operation of a first manufacturing process, can be input into the digital replica. The digital replica can include a physics-based model of the manufacturing process and / or facility. The physics-based model can enable what-if scenarios in which module / subsystem consumption is estimated using a fully physics-based model or a reduced-order model (e.g., lamp heating variations conceptualized before a subsystem, such as an infrared-based lamp heating subsystem, is developed). In some embodiments, the digital replica can include other models, such as statistical models, to determine the physical conditions of the manufacturing process or facility (e.g., heating losses, energy consumption, etc., performed by gases exiting the exhaust and / or foreline).

[0020] In some embodiments, eco-efficiency characterization can be integrated across multiple manufacturing processes. For example, cumulative consumption per device or process component can be calculated across various devices and processes to compile a cumulative eco-efficiency. Additionally or alternatively, auxiliary or support equipment (e.g., sub-fabrication equipment), such as equipment shared across multiple manufacturing facilities, can be characterized. For example, devices and / or equipment such as pumps, abatement devices, heater jackets, filtration systems, or other devices not used to directly process substrates can also be monitored and characterized for eco-efficiency.

[0021] In some embodiments, modifications to a manufacturing process (e.g., a subset of processes or multiple processes) may be determined based on the environmental resource usage data or the eco-efficiency characterization. For example, the environmental resource usage data may be used as input to a machine learning model. One or more outputs from the machine learning model may be obtained that indicate the modifications to the manufacturing process and, in some embodiments, a confidence level that the modifications satisfy a threshold condition. The modifications to the manufacturing process may be associated with improving the eco-efficiency (e.g., reducing environmental resource consumption and / or environmental impact) of the selected manufacturing process.

[0022] In some embodiments, determining one or more modifications to the manufacturing process may be associated with an optimization procedure for the manufacturing process. The system and / or methodology may determine multiple modifications to apply to the manufacturing process to meet predetermined environmental optimization requirements. For example, local regulations may impose limits on the level of use of certain resources (e.g., electricity, water, etc.) or incentivize lower consumption practices through different reward mechanisms. The eco-efficiency systems and methodologies described herein (e.g., real-time dashboards as monitoring features) may be readily used to prepare the necessary reports as evidence of compliance, and the eco-efficiency optimization features may be used to realize savings for the manufacturing system.

[0023] In some embodiments, the compliance report may include a report based on a generally accepted code and / or standard, such as the Semiconductor Equipment and Materials Association (SEMI) published in Semiconductor Facility System Guidelines for Energy, Electricity, and Productive Maintenance for Semiconductor Manufacturing Facilities (SEMI S23-0813). For example, SEMI S23-0813 provides the energy conversion factor (ECF) (e.g., energy consumption per unit of flow rate) of key utilities. The ECF may estimate the energy consumption of utilities and be used to estimate energy savings in semiconductor manufacturing facilities.

[0024] In some embodiments, eco-efficiency is based on resource consumption, such as, for example, energy consumption, gas consumption (such as hydrogen, nitrogen, chemicals used for etching or depositing thin films, CDA (clean dry air)), and / or water consumption (such as process cooling water (PCW), deionized water (DIW), and ultrapure water (UPW)). However, in some embodiments, eco-efficiency is based on life cycle data of components associated with the manufacturing equipment. For example, the environmental resource consumption and / or environmental impact associated with the eco-efficiency characterization may be associated with replacement or maintenance procedures for consumable parts of the manufacturing equipment. Modifications may be associated with maintenance procedures for consumable parts of the manufacturing equipment.

[0025] In some embodiments, characterizing and optimizing a manufacturing process may include a recipe builder methodology. The recipe builder methodology may include dynamically calculating the resource consumption and / or environmental impact of individual manufacturing steps as part of the recipe creation and / or modification process. A user may be able to add, delete, and / or modify various combinations, subcombinations, and / or sequences of process steps and / or manufacturing equipment to perform the processing steps. The modified recipe may be analyzed to determine the environmental efficiency of the modified recipe. Performing a process may include, for example, processing wafers, transporting wafers, auxiliary / support equipment that enables the process steps, and / or other functions associated with the manufacturing process.

[0026] 1 is a block diagram illustrating an example system architecture 100 in which embodiments of the present disclosure may operate. As shown in FIG. 1, the system architecture 100 includes a manufacturing system 102, a data store 112, a server 120, a client device 150, and / or a machine learning system 170. The machine learning system 170 may be part of the server 120. In some embodiments, one or more components of the machine learning system 170 may be fully or partially integrated into the client device 150. The manufacturing system 102, the data store 112, the server 120, the client device 150, and the machine learning system 170 may each be hosted by one or more computing devices, including a server computer, a desktop computer, a laptop computer, a tablet computer, a notebook computer, a personal digital assistant (PDA), a mobile communication device, a mobile phone, a handheld computer, or a similar computing device.

[0027] The manufacturing system 102, the data store 112, the server 120, the client device 150, and the machine learning system 170 may be coupled to each other via a network (e.g., to implement the methodologies described herein). In some embodiments, the network 160 is a private network that provides each element of the system architecture 100 with access to each other and to other privately available computing devices. The network 160 may include one or more wide area networks (WANs), local area networks (LANs), wired networks (e.g., Ethernet networks), wireless networks (e.g., 802.11 networks or Wi-Fi networks), cellular networks (e.g., Long Term Evolution (LTE) networks), cloud networks, cloud services, routers, hubs, switches, server computers, and / or any combination thereof. Alternatively or additionally, any of the elements of the system architecture 100 may be integrated together or separately coupled without the use of the network 160.

[0028] Client device 150 may be or include any personal computer (PC), laptop, mobile phone, tablet computer, netbook computer, network-connected television (“smart TV”), network-connected media player (e.g., Blu-ray player), set-top box, over-the-top (OOT) streaming device, operator box, etc. Client device 150 may include a browser 152, applications 154, and / or other tools as described and as implemented by other systems in system architecture 100. In some embodiments, client device 150 may be capable of accessing manufacturing system 102, data store 112, server 120, and / or machine learning system 170 and communicating (e.g., sending and / or receiving) one or more environmental resource consumption (e.g., environmental resource consumption) and / or eco-efficiency, including environmental impact, and / or inputs and outputs of various process tools (e.g., component integration tool 122, digital replica tool 124, optimization tool 126, recipe builder tool 128, resource consumption tool 130, etc.) at various stages of processing of system architecture 100 as described herein.

[0029] 1, the manufacturing system 102 includes tooling 104, tool controller 106, process recipe 108, and sensor 110. Tooling 104 may be any combination of ion implanters, etch reactors (e.g., processing chambers), photolithography devices, deposition devices (e.g., for performing chemical vapor deposition (CVD), physical vapor deposition (PVD), ion-assisted deposition (IAD), etc.), or any other combination of manufacturing devices.

[0030] The process recipe 108, also referred to as a manufacturing recipe or manufacturing process instructions, includes a sequence of machine operations involving process embodiments that, when applied in a specified order, create a manufacturing sample (e.g., a substrate or wafer having predetermined properties or meeting predetermined specifications). In some embodiments, the process recipe is stored in a data store or, alternatively or additionally, is stored in a manner that generates a table of data representing the steps or operations of the manufacturing process. Each step may store the known environmental efficiency of a given process step. Alternatively or additionally, each process step may store parameters that indicate the physical conditions required by the process step (e.g., target pressure, temperature, exhaust gases, energy throughput, and the like).

[0031] The equipment controller 106 may include software and / or hardware components capable of executing the steps of the process recipe 108. The equipment controller 106 may monitor the manufacturing process through sensors 110. The sensors 110 may measure process parameters to determine whether process criteria are met. The process criteria may be associated with process parameter value windows (e.g., as described in connection with FIG. 5 ). The sensors 110 may include various sensors that may be used to measure consumption (e.g., power, current, etc.) (explicitly or as an indicator). The sensors 110 may include physical sensors, Internet of Things (IoT), and / or virtual sensors (e.g., sensors that are not physical sensors but are based on virtual measurements based on models that estimate parameter values).

[0032] Additionally or alternatively, the equipment controller 106 may monitor eco-efficiency by measuring resource consumption (e.g., emissions, energy consumption, process material consumption, etc.) of various process steps. In some embodiments, the equipment controller 106 determines the eco-efficiency of the associated machine equipment 104. The equipment controller 106 may also adjust settings associated with the manufacturing equipment 104 based on the determined eco-efficiency model (e.g., including determined modifications to the process recipe 108) to optimize the eco-efficiency of the equipment 104 given the current manufacturing conditions.

[0033] In one embodiment, the equipment controller 106 may include a main memory (e.g., read-only memory (ROM), flash memory, dynamic random access memory (DRAM), static random access memory (SRAM), etc.) and / or a secondary memory (e.g., a data store device such as a disk drive (e.g., data store 112 or cloud data). The main memory and / or secondary memory may store instructions for performing various types of manufacturing processes (e.g., process recipes 108).

[0034] In one embodiment, the equipment controller 106 may determine an actual eco-efficiency characteristic associated with the manufacturing facility 104 based on first utility usage data associated with the manufacturing facility 104 and first usage data associated with the manufacturing facility 104. The first utility usage data and the first usage data may be determined, for example, by the equipment controller 106. In another embodiment, the first utility usage data and the first usage data are received from an external source (e.g., the server 120, a cloud service, and / or a cloud data store). The equipment controller 106 may compare the actual eco-efficiency characteristic to a first eco-efficiency characteristic associated with the manufacturing facility 104 (e.g., a first estimated eco-efficiency characteristic). The eco-efficiency characteristic may differ when usage and utilization data values ​​that differ from the actual values ​​associated with the operating manufacturing facility 104 are used to calculate the first eco-efficiency characteristic.

[0035] In one embodiment, the equipment controller 106 may determine that the first eco-efficiency characterization is more eco-efficient than the actual eco-efficiency characterization, indicating that it may be possible to adjust settings for the manufacturing equipment 104 to better optimize the manufacturing equipment 104 for eco-efficiency. In some embodiments, the manufacturing equipment 104 may control and adjust sub-component settings to better optimize eco-efficiency.

[0036] The equipment controller 106 may also determine, based on the actual usage data, the actual utilization data, and the eco-efficiency characterization, that the actual usage data or utilization data is not the same as the usage data and utilization data associated with the first eco-efficiency characterization. This may be the case when nominal or estimated data values ​​are used to determine the first eco-efficiency characterization and different actual recorded data values ​​are used while the manufacturing equipment 104 is operating. In such a scenario, adjustments to one or more settings associated with the manufacturing equipment 104 may be beneficial to optimize the eco-efficiency of the manufacturing equipment.

[0037] The data store 112 may be another type of component or device capable of storing data, such as a memory (e.g., random access memory), a drive (e.g., a hard drive, a flash drive), a database system, or a store provided by a cloud server and / or processor. The data store 112 may store one or more historical sensor data. The data store 112 may store one or more eco-efficiency data 114 (e.g., including past and / or current eco-efficiency data), sensor and process recipe data 116 (e.g., including past and / or current sensor and process recipe data 116), and correction and optimization data (e.g., past and / or current correction and optimization data 118). The sensor and process recipe data 116 may include various process steps, process parameter windows, alternative process steps, process queuing instructions, etc. for performing multiple processes on overlapping manufacturing equipment. The sensor and process recipe data 116 may be linked or otherwise associated with the eco-efficiency data 114 to track eco-efficiency across various process steps, recipes, etc. The modification and optimization data 118 may include historical modifications made to previous process recipes (individual process steps or multiple process recipes in combination) and the associated eco-efficiency changes resulting from the modifications.

[0038] The eco-efficiency data 114 may include various consumed resources used in the eco-efficiency characterization. In one embodiment, the eco-efficiency data 114 incorporates one or more of water usage, emissions, electrical energy usage, and any combination thereof. In other embodiments, the eco-efficiency data 114 may include other categories of resource consumption, such as gas usage, heavy metal usage, and eutrophication potential.

[0039] Server 120 may include a component integration tool 122, a digital replica tool 124, an optimization tool 126, a recipe builder tool 128, and / or a resource consumption tool 130. Component integration tool 122 may determine cumulative consumption per device (e.g., per individual manufacturing facility). The various tools of server 120 may communicate data between each other to perform their respective functions as described herein.

[0040] The component integration tool 122 may receive manufacturing data (e.g., recipes, recipe selections, manufacturing equipment, inter-recipe and intra-recipe processes, etc.) and perform eco-efficiency analysis across various sections of the data. In some embodiments, the component integration tool 122 may determine eco-efficiency characterizations across multiple process steps from individual process recipes. For example, the component integration tool 122 may determine eco-efficiency characterizations across all steps of a recipe from start to finish. In another example, a process selection may be used to determine the eco-efficiency of a subset of manufacturing process steps.

[0041] In another embodiment, the component integration tool 122 may perform an inter-recipe process eco-efficiency characterization. For example, the eco-efficiency characterization may be associated with a manufacturing device (e.g., of the manufacturing system 102) that performs multiple different process steps from multiple different manufacturing processes (e.g., the process recipe 108). In another example, the order of various process steps (e.g., within a recipe or between recipes) may affect the overall eco-efficiency. The component integration tool 122 may perform an overall eco-efficiency characterization across systems and / or process sequences of the manufacturing device. For example, the component integration tool 122 may perform an eco-efficiency comparison between sub-components (e.g., multiple processing chambers) that perform similar functions.

[0042] In an illustrative example, each process step may be performed by a processing chamber, such as epitaxial deposition or etching. Each of these is performed using a process recipe. There may be many different process recipes for performing a process such as epitaxial deposition. For example, a process recipe may include multiple steps such as 1) purging the chamber, 2) pumping, 3) flowing gases, 4) heating the chamber, etc. These steps may be associated with one or more process recipes.

[0043] In another embodiment, component integration tool 122 may perform eco-efficiency characterization, including eco-efficiency of auxiliary equipment. Auxiliary equipment may include equipment not directly used for manufacturing but that assists in executing various process recipes. For example, auxiliary equipment may include a substrate transport system designed to move wafers between various manufacturing devices. In another example, auxiliary equipment may include heat sinks, shared exhaust ports, power delivery systems, etc. Component integration tool 122 may consider auxiliary device resource consumption and combine auxiliary device resource consumption with manufacturing resource consumption to determine resource consumption for a process recipe (e.g., a subset or full recipe) or combination of recipes (e.g., a subset or full recipe).

[0044] In another embodiment, component integration tool 122 may perform an eco-efficiency characterization that considers a sequence of processes or recipes. For example, performing process step A followed by process step B may result in a first resource consumption, while performing process step B followed by process step A may result in a second resource consumption that differs from the first resource consumption. Component integration tool 122 integrates eco-efficiency across multiple equipment and / or process steps and considers the sequence of process steps for a process recipe (e.g., a subset or full recipe) or combination of recipes (e.g., a subset or full recipe).

[0045] In some embodiments, there is a different manufacturing facility for each process step. For example, a film on a wafer may have multiple layers. A first machine may perform a first operation (e.g., deposition), a second machine may perform a second operation (e.g., etch), a third machine may perform a third operation (e.g., deposition), and so on. The component integration tool 122 may instruct a resource consumption tracker to track multiple processing steps across multiple machines and generate a data archival report. As previously mentioned, the consumption report may be plotted for a selection of process recipes, including the life of the wafer from start to finish.

[0046] The digital replica tool 124 receives manufacturing data from the manufacturing system 102 and / or the client device 150 and generates a digital replica associated with the manufacturing data. The manufacturing data may include the selection of process steps for the tooling 104 and the process recipe 108. The digital replica tool 124 generates a digital twin of the manufacturing system's physical system architecture or a virtual input system (e.g., generated by a user on the client device 150).

[0047] The digital replica generated by the digital replica tool 124 may include one of a physical model, a statistical model, and / or a hybrid model. The physical model may include physics-based constraints and control algorithms designed to estimate the physical conditions of the input manufacturing data (e.g., exhaust gas temperature, power delivery requirements, and / or other conditions indicative of the physical environment associated with environmental resource consumption). For example, a user may create a process recipe on the client device 150. The process recipe may include parameters for a process or recipe and instructions for using mechanical equipment in a specific manner. The digital replica tool 124 may use this manufacturing data to determine the physical constraints of the system (e.g., operating temperature, pressure, exhaust gas parameters, etc.). For example, the physical model may identify the physical conditions of the system based on the chamber hardware configuration (e.g., whether Type A or Type B equipment materials are used) and / or recipe parameters. In another example, the physical conditions may be determined from associated mechanical equipment components that affect heat loss to water, air, and / or heating, ventilation, and air conditioning (HVAC) equipment. The digital replica tool 124 may cooperate with other tools (e.g., the component integration tool 122 and / or the resource consumption tool 130) to predict the eco-efficiency characterization of the received manufacturing data. It should be noted that the digital replica tool 124 may predict the eco-efficiency of a manufacturing process and selected manufacturing equipment without receiving empirical data from performing the process recipe with the manufacturing equipment 104. Thus, the digital replica of the manufacturing equipment may be used to predict the eco-efficiency of an equipment design and / or a process recipe without actually building the particular equipment design or running the particular process recipe.

[0048] In some embodiments, the physical models used by the digital replica tool 124 may include fluid flow modeling, gas flow and / or consumption modeling, chemistry-based modeling, heat transfer modeling, electrical energy consumption modeling, plasma modeling, and the like.

[0049] In some embodiments, the digital replica tool 124 may predict the eco-efficiency of the manufacturing data using statistical modeling. Statistical models may be used to process the manufacturing data based on previously processed historical eco-efficiency data (e.g., eco-efficiency data 114) using statistical operations to validate, predict, and / or transform the manufacturing data. In some embodiments, statistical models are generated using statistical process control (SPC) analysis to determine control limits for the data and identify the data as more or less reliable based on those control limits. In some embodiments, statistical models are associated with univariate and / or multivariate data analysis. For example, various parameters may be analyzed using statistical models to determine patterns and correlations through statistical processes (e.g., range, minimum, maximum, quartiles, variance, standard deviation, etc.). In another example, relationships between multiple variables may be identified using regression analysis, path analysis, factor analysis, multivariate statistical process control (MCSPC), and / or multivariate analysis of variance (MANOVA).

[0050] The optimization tool 126 may receive the process recipe 108 and tooling 104 selections and may identify modifications to the selections to improve eco-efficiency (e.g., reduce resource consumption, resource cost consumption, and / or environmental impact (e.g., gas or particulate species entering the atmosphere)). The optimization tool 126 may incorporate the use of a machine learning model (e.g., model 190 of machine learning system 170). The machine learning model may receive the process recipe and / or tooling selections as input and determine one or more modifications to the selections that, when implemented by the manufacturing system 102, improve the overall eco-efficiency of the selections. In some embodiments, the machine learning model may use a digital replica tool to generate synthetic manufacturing data for training. Alternatively or additionally, the machine learning model may use historical data (e.g., eco-efficiency data 114, sensor and process recipe data 116, and / or modification and optimization data 118) to train the machine learning model.

[0051] The modifications identified by the optimization tool 126 may include changing process steps, changing the order of processes, changing parameters performed by a piece of tooling, changing the interaction (e.g., order, simultaneous operations, delay times, etc.) between a first process recipe and a second process recipe, etc. In some embodiments, the optimization tool 126 may send instructions to the manufacturing system 102 to perform the optimization directly. However, in other embodiments, the optimization tool may display the modifications on a graphical user interface (GUI) for an operator to manipulate. For example, the digital replica tool 124 may send one or more modifications to the client device 150 for display within the browser 152 and / or application 154.

[0052] In some embodiments, optimization tool 126 may adjust hyperparameters of the digital twin model generated by digital replica tool 124. As discussed in later embodiments, optimization tool 126 may incorporate reinforcement learning and / or deep learning by performing simulated modifications to the digital replica and evaluating eco-efficiency results output from the digital replica.

[0053] In some embodiments, optimization tool 126 may perform an eco-efficiency characterization and an optimization that prioritizes one or more types of environmental resources. For example, as described above, the eco-efficiency characterization may be based on various resource consumptions, such as water usage, gas usage, and energy usage. Optimization tool 126 may perform an optimization that prioritizes a first resource consumption (e.g., water usage) over a second resource consumption (e.g., gas usage). In some embodiments, optimization tool 126 may perform an optimization using a weighted priority system. For example, when optimizing eco-efficiency and / or identifying eco-efficiency modifications to a manufacturing process, one or more resource consumptions may be assigned a weight that indicates an optimization priority for the associated per-unit resource consumption.

[0054] The recipe builder tool 128 may receive a manufacturing process and / or equipment selection and dynamically determine and predict eco-efficiency for each step after each addition, deletion, and / or modification to the virtual manufacturing process and / or equipment selection. The recipe builder tool 128 may use other tools (e.g., the component integration tool 122, the digital replica tool 124, the optimization tool 126, and the resource consumption tool 130) to dynamically update the determined eco-efficiency when the manufacturing recipe is updated. For example, a user may create a manufacturing recipe. The recipe builder tool 128 may output a current eco-efficiency for the current iteration of the process recipe. The recipe builder tool 128 may receive modifications to the current iteration that updated the process recipe. The recipe builder tool 128 may output an updated eco-efficiency characterization.

[0055] The resource consumption tool 130 may track various resource consumptions. For example, as described above, environmental characterization may be based on broader resources such as energy consumption, gas emissions, and water usage. However, the resource consumption tool 130 may track resource consumption more specifically. In some embodiments, a selection of a process recipe and / or manufacturing equipment is received by the resource consumption tool 130. The resource consumption tool 130 may determine life cycle data for components associated with the selection of manufacturing equipment and / or process recipe. For example, manufacturing equipment wears out with use and requires corrective action, such as replacing and / or repairing components. This corrective action is also associated with environmental consumption (e.g., resource consumption for implementing the corrective action). The resource consumption tool 130 may track component life data individually and provide per-unit environmental resource consumption and / or environmental impact based on future corrective actions expected to be implemented.

[0056] In some embodiments, the digital twin can be used to estimate the lifespan of certain consumables associated with a manufacturing process step. The lifespan data can be used to estimate lifespan duration and predict the next corrective steps to be taken in response to the predicted lifespan. For example, the lifespan data can be used to maintain optimized eco-efficiency performance by proactively notifying the supply chain about replacement part orders.

[0057] In some embodiments, the environmental resource usage data determined by other tools of the server may include environmental resource consumption and / or environmental impact associated with one of a replacement procedure or a maintenance procedure for a consumable part of the first manufacturing equipment. In some embodiments, the optimization tool 126 may determine modifications to the manufacturing process, which may include implementing corrective actions associated with components of the machine equipment (e.g., machine equipment 104).

[0058] In some embodiments, machine learning system 170 further includes server machine 172, server machine 180, and / or server machine 192. Server machine 172 includes dataset generator 174 that can generate datasets (e.g., a set of data inputs and a set of target outputs) for training, validating, and / or testing machine learning model 190. Some operations of dataset generator 174 are described in more detail below with respect to FIGS. 7A-C.

[0059] Server machine 180 includes a training engine 182, a validation engine 184, and / or a test engine 186. An engine (e.g., training engine 182, validation engine 184, and / or test engine 186) may refer to hardware (e.g., circuitry, dedicated logic, programmable logic, microcode, processing device, etc.), software (e.g., instructions executing on a processing device, general-purpose computer system, or dedicated machine), firmware, microcode, or a combination thereof. Training engine 182 may be capable of training machine learning models 190 using one or more sets of features associated with training sets from dataset generator 174. Training engine 182 may generate one or more trained machine learning models 190, each of which may be trained based on a distinct set of features of the training set and / or a distinct set of labels of the training set. For example, a first trained machine learning model may have been trained using resource consumption data output by the digital replica tool 124, a second trained machine learning model may have been trained using historical eco-efficiency data (e.g., eco-efficiency data 114), etc.

[0060] The validation engine 184 may be capable of validating the trained machine learning model 190 using the validation set from the dataset generator 174. The test engine 186 may be capable of testing the trained machine learning model 190 using the test set from the dataset generator 174.

[0061] Machine learning model 190 may refer to one or more trained machine learning models created by training engine 182 using a training set including data inputs and, in some embodiments, corresponding target outputs (e.g., correct answers for each training input). Patterns in a dataset that cluster data inputs and / or map data inputs to target outputs (correct answers) may be discovered, and machine learning model 190 is provided with the mappings and / or learns mappings that capture these patterns. Machine learning model 190 may include artificial neural networks, deep neural networks, convolutional neural networks, recurrent neural networks (e.g., long short-term memory (LSTM) networks, convLSTM networks, etc.), and / or other types of neural networks. Machine learning model 190 may additionally or alternatively include other types of machine learning models, such as those using one or more of linear regression, Gaussian regression, random forests, support vector machines, etc.

[0062] The modification identification component 194 may provide the current data to the trained machine learning model 190 and may execute the trained machine learning model 190 on the input to obtain one or more outputs. The modification identification component 194 may be able to make decisions and / or perform actions from the output of the trained machine learning model 190. The ML model output may include confidence data indicating a degree of confidence that the ML model output (e.g., modifications and optimization parameters) corresponds to a modification that, when applied, improves the overall eco-efficiency of the manufacturing process and / or manufacturing equipment selection. The modification identification component 194, in some embodiments, may implement a process recipe modification based on the ML model output. The modification identification component 194 may provide the ML model output to one or more tools of the server 120.

[0063] The confidence data may include or indicate a confidence that the ML model output is correct (e.g., the ML model output corresponds to a known label associated with a training data item). In one example, the confidence is a real number between 0 and 1, inclusive, where 0 indicates no confidence that the ML model output is correct and 1 indicates absolute confidence that the ML model output is correct. In response to confidence data indicating a confidence below a threshold level across a predetermined number of instances (e.g., a percentage of instances, a frequency of instances, a total number of instances, etc.), the server 120 may cause the trained machine learning model 190 to be retrained.

[0064] For purposes of illustration and not limitation, aspects of the present disclosure describe training a machine learning model that uses process recipe data and inputs current selections of manufacturing processes and / or manufacturing equipment into the trained machine learning model to determine ML model outputs (process modifications and optimization parameters, such as target eco-efficiencies for specific resource consumption). In other embodiments, heuristic or rule-based models are used to determine the outputs (e.g., without using a trained machine learning model).

[0065] In some embodiments, the functionality of manufacturing system 102, client device 150, machine learning system 170, data store 112, and / or server 120 may be provided by fewer machines. For example, in some embodiments, server machines 172 and 180 may be combined into a single machine, while in some embodiments, server machine 172, server machine 180, and server machine 192 may be combined into a single machine. In some embodiments, server 120, manufacturing system 102, and client device 150 may be combined into a single machine.

[0066] In general, functionality described as being performed by manufacturing system 102, client device 150, and / or machine learning system 170 in one embodiment may also be performed on server 120 in other embodiments, where appropriate. In addition, functionality attributed to a particular component may be performed by a different component or multiple components operating together. For example, in some embodiments, server 120 may receive manufacturing data and perform machine learning operations. In another example, client device 150 may perform manufacturing data processing based on output from a trained machine learning model.

[0067] Additionally, the functionality of a particular component may be performed by a different component or multiple components working together. One or more of the server 120, the manufacturing system 102, or the machine learning system 170 may be accessed as a service offered to other systems or devices through an appropriate application programming interface (API).

[0068] In embodiments, a "user" may be represented as a single individual. However, other embodiments of the present disclosure encompass a "user" that is an entity controlled by multiple users and / or automated sources. For example, a set of individual users that are aggregated as a group of administrators may be considered a "user."

[0069] 2 is a block diagram illustrating an eco-efficiency sustainability system architecture 200 in which embodiments of the present disclosure may operate. The system architecture 200 includes a selection of processing tools 202 having one or more subcomponents 204 (e.g., processing chambers). As previously described, the processing tools 202 include various manufacturing tools used to process substrates. In line 206, sensors measure and transmit manufacturing data (e.g., energy consumption sensor data, gas and water consumption data, etc.) to a common ground architecture 208. The common ground architecture 208 may include one or more control algorithms configured to perform manufacturing process steps and manage process parameters (e.g., critical process parameters, machine equipment diagnostic parameters, parameters otherwise indicative of the manufacturing process).

[0070] The common ground architecture 208 may transmit sensor data (e.g., from wired sensors and / or wireless sensors, such as Internet of Things (IoT) sensors) to a data management algorithm (e.g., an integration algorithm 210). The integration algorithm 210 may analyze manufacturing data received from the processing tools 202 to select a portion of the data for performing an eco-efficiency characterization. The integration algorithm 210 extracts data to perform a cumulative eco-efficiency characterization for a selection of manufacturing process steps and / or manufacturing equipment. The selected data may be used in conjunction with a physics-based model 214 to determine the physical conditions of the processing tools 202 (e.g., of each subcomponent 204). The data may be combined (e.g., at line 218) with scheduling information from an onboard sequencer and / or planner or from an operator. The scheduling information may include data indicative of upcoming recipes, tool idle states, maintenance, etc.

[0071] The selection of manufacturing data combined with schedule data is input into physics-based model 214. In some embodiments, the physics-based model is a machine model. The machine model examines the mechanics of the individual data points of the manufacturing data and the schedule information, and the manner in which the individual data points are combined, to determine a physical / mechanical representation of the data combination. In some embodiments, the machine model may include processing the data to determine a prediction of resource consumption. For example, the machine model may process the manufacturing data to determine a resource consumption (e.g., water, energy, gas, etc.) prediction and / or an environmental impact (e.g., gas or particulate species entering the atmosphere). The machine model may be generated using historical manufacturing data and later used against current data to determine a prediction.

[0072] In some embodiments, the physics-based model 214 may incorporate various physical relationships, such as thermodynamics, fluid mechanics, energy conservation, gas laws, mechanical systems, energy conservation, transport, and delivery. For example, a processing tool may include a cooling water flow to a portion of a manufacturing equipment device to perform a cooling process. The physics model may combine fluid mechanics with heat transfer to determine a model for converting raw manufacturing data into system process data that can be characterized for its eco-efficiency. In some embodiments, the physics model may be used to determine whether a threshold resource consumption condition is met. By the same example, the physics model may be used to determine the flow rate of a fluid within a subcomponent, and therefore the heat transfer rate. If this heat transfer rate falls below the threshold rate, additional energy may be lost to emissions. Thus, the physics model may determine that the fluid flow rate is operating below a preferred desired flow rate level to maintain a desired level of eco-efficiency.

[0073] In some embodiments, the physics-based model 214 incorporates auxiliary or peripheral equipment operating resource consumption, such as the energy consumption of powering processing devices to provide control algorithms to the processing tool 202 (e.g., using the common ground architecture 208). The auxiliary equipment may not be located in close proximity to the manufacturing equipment or directly associated with a single manufacturing process, but may be allocated as a contribution to various manufacturing process steps (or individual manufacturing processes) using the physics-based model 214.

[0074] In some embodiments, in addition to or as an alternative to using physical models, statistical models are used on manufacturing data. Statistical models can be used to process data based on statistical operations to validate, predict, and / or transform manufacturing data. In some embodiments, statistical models are generated using statistical process control (SPC) analysis to determine control limits for the data and identify data as more or less reliable based on those control limits. In some embodiments, statistical models are associated with univariate and / or multivariate data analysis. For example, various parameters can be analyzed using statistical models to determine patterns and correlations through statistical processes (e.g., range, minimum, maximum, quartiles, variance, standard deviation, etc.). In another example, relationships between multiple variables can be identified using regression analysis, path analysis, factor analysis, multivariate statistical process control (MCSPC), and / or multivariate analysis of variance (MANOVA).

[0075] In some embodiments, system architecture 200 includes an adaptive optimization algorithm 216. The adaptive optimization algorithm 216 cooperates with physics-based model 214 to determine modifications to the manufacturing process and / or the selection of manufacturing equipment to perform the associated processes. In some embodiments, the adaptive optimization algorithm outputs automatic optimization commands to control software (e.g., at line 224). In other embodiments, the adaptive optimization algorithm may output suggestions to an operator to optimize performance (e.g., at line 220). In some embodiments, the adaptive optimization algorithm outputs automatic optimization commands for hardware components (e.g., at line 222).

[0076] In some embodiments, the adaptive optimization algorithm 216 uses a machine learning model to determine modifications to the manufacturing process and / or manufacturing equipment. The machine learning model may be a trained machine learning model (e.g., trained and executed using methods 700A-C). As discussed in further embodiments, the machine learning model may operate in conjunction with a physics-based model to determine modifications to the manufacturing process and / or equipment received as input.

[0077] The system architecture 200 may include an integrated dashboard GUI 212. The integrated dashboard GUI may be designed to display relevant manufacturing data (e.g., sensor data, machine diagnostics, machine status, manufacturing process status, etc.). In some embodiments, the integrated dashboard GUI includes a method for receiving input from a user. For example, a user may input manufacturing data to generate a recipe (e.g., using the recipe builder tool 128). This additional manufacturing data may be used as input to one or more of the physics-based models and adaptive optimization algorithms 216.

[0078] 3 depicts a flow diagram of an example methodology 300 for monitoring, sustaining, and / or optimizing a manufacturing process. The example methodology can be divided into two parts: first, training a machine learning model 324, and second, implementing a manufacturing process 324. The example methodology 300, in one embodiment, includes a machine learning model 302, tool software 306, tool hardware 308, and a physics model 312.

[0079] In some embodiments, the machine learning model receives a selection of a manufacturing process and / or manufacturing equipment and outputs one or more modifications to the manufacturing process and / or manufacturing equipment to improve environmental efficiency (e.g., reduce resource consumption). In some embodiments, the physical model 312 is used to generate simulated training / validation data 320 (e.g., using method 700A). In response to the received simulated training / validation data 320, the machine learning model 302 generates simulated modifications 318 that can be returned to the physical model 312 for validation. The machine learning model 302 is trained on various simulated and / or actual training / validation data 320. Once trained, the machine learning model 302 may receive a selection of an empirical manufacturing system and / or process recipe to be implemented by the system. The machine learning model 302 outputs manufacturing process instructions and / or modifications 304 to an equipment controller (e.g., equipment controller 106) implementing the tool software 306. These modifications may improve environmental efficiency. The tool software 306 provides manufacturing process instructions 314 to the tool hardware 308. The tool hardware 308 performs the manufacturing process and includes sensors that report sensor data back to the equipment controller that implements the tool software 306.

[0080] In some embodiments, the equipment controller identifies one or more physical conditions of the tool hardware as violating a threshold condition (e.g., high temperature, overpressure, gas leak, power shortage, etc.). The equipment controller may modify the manufacturing process instructions to remedy the violated threshold condition (e.g., based on output from the machine learning model 302).

[0081] The manufacturing system, including the tool hardware 308, reports back the empirical training / validation 310 to the physics model 312. The physics model may then generate and update simulated training / validation data 320, which may be updated and used for further training of the machine learning model.

[0082] In some embodiments, the physics model 312 generates simulated training / validation data, while in other embodiments, the physics model outputs modifications to the manufacturing process. In such embodiments, a machine learning model may be used as an optimization model to adjust hyperparameters (e.g., manufacturing data parameters) to identify modifications to further optimize the manufacturing process. For example, the manufacturing process may be used as input to the physics model 312. The machine learning model may then process the output of the physics model 312 to identify possible changes to the manufacturing process (i.e., hyperparameters). The identified changes may be run against the physics model to determine corresponding updated eco-efficiencies. This may be repeated in an iterative process to fine-tune the equipment design and / or recipe design. In examples, the optimization model may be generated and / or implemented using an instance of the Broyden-Fletcher-Goldfarb-Shanno (BFGS) algorithm, a conjugate gradient (CG) algorithm, an instance of the Nelder-Mead algorithm, and / or a model predictive control (MPC) algorithm.

[0083] 4 depicts an example digital replica 400 according to some embodiments of the present disclosure. Digital replica 400 may include a digital twin of a selection of a manufacturing system, for example, a digital replication of the manufacturing system including the same chambers, valves, gas supply lines, materials, chamber components, etc. Digital replica 400 may receive manufacturing equipment process data (e.g., sensor data) 404A-C and a process recipe 404D as inputs and output the physical conditions 406 of the manufacturing system. In some embodiments, digital replica 400 includes physics-based models that may incorporate various physical relationships, such as thermodynamics, fluid dynamics, energy conservation, gas laws, mechanical systems, energy conservation, transport, and delivery.

[0084] For example, as seen in FIG. 4 , the digital replica receives as input a first gas flow of a first gas 404A, a second gas flow of a second gas 404B, and a third gas flow of a third gas 404C, as well as a first process recipe 404D. The digital replica uses a physics-based model to estimate the amount of energy leaving the chamber due to the gas flows. For example, the model determines the temperature of the exhaust gases and the total energy flow through the exhaust holes. In another example, the same digital replica 400 can output eco-efficiency optimization modifications, such as a different hardware configuration for the chamber (e.g., whether to use a first ray type A or a second ray type B). The digital replica can identify relevant parts of the system that affect heat loss to water, air, and HVAC and identify suggested optimizations to improve energy conservation.

[0085] FIG. 5 is an exemplary illustration of operational parameter constraints 500 for manufacturing process steps according to some embodiments of the present disclosure. Various manufacturing process steps may include operational parameter constraints 500 that indicate process parameter windows 510, or sets of values ​​(e.g., combinations of values), for a corresponding set of parameters that, when satisfied, result in meeting a threshold condition (e.g., a minimum quality condition). For example, the process parameter window 510 may include a first parameter 502 (e.g., a first flow rate of a first gas) and a second parameter 504 (e.g., a temperature of the gas). To perform a manufacturing process and meet a threshold condition (e.g., a minimum quality specification, a statistical process control (SPC) limit, a specification limit, etc.), the process parameter value window 510 is determined to identify a parameter value combination that results in a product that is likely to meet the threshold condition. As shown in FIG. 5, the process parameter window 510 includes a lower limit 506A and an upper limit 506A for the first parameter 502 and a lower limit 508B and an upper limit 508A for the second parameter.

[0086] The optimization determined by the manufacturing process system (e.g., using the adaptive optimization algorithm 216 and / or the physics-based model 214) may include determining an environmentally optimized process parameter window 512 within the process parameter window 510 that causes the manufacturing operation to consume a reduced amount of resources compared to process parameter values ​​outside the environmentally optimized process parameter window 512.

[0087] 5 depicts a simplified process parameter window 510 and an environmentally optimized process parameter window 512 that depend on only two parameters 502, 504. Both the process parameter window 510 and the environmentally optimized process parameter window 512 form a simple rectangle. The process parameter windows may include more than two parameters and may include more diverse parameter dependencies. For example, nonlinear, physics-based relationships between parameters may result in nonlinear process parameter windows and environmentally optimized process parameter windows.

[0088] 6 and 7A-C depict flow diagrams illustrating example methods 600, 700A-C related to training and / or using machine learning models in connection with environmental resource consumption and / or environmental impact of manufacturing processes, according to some embodiments of the present disclosure. For simplicity of explanation, methods 600, 700A-C are depicted and described as a series of acts. However, acts in accordance with the present disclosure may occur in various orders and / or simultaneously, and with other acts not presented and described herein. Furthermore, not all of the illustrated acts may be performed to implement methods 600, 700A-C in accordance with the disclosed subject matter. Additionally, those skilled in the art will understand and appreciate that methods 600, 700A-C may alternatively be represented as a series of interrelated states via state diagrams or events.

[0089] 6 is an exemplary illustration of a method 600 for identifying modifications to a manufacturing process according to some embodiments of the present disclosure. Method 600 is implemented by processing logic, which may include hardware (circuitry, dedicated logic, etc.), software (e.g., running on a general-purpose computer system or a dedicated machine), or any combination thereof. In one embodiment, the method is implemented using server 120 and trained machine learning model 190 of FIG. 1 , although in some other embodiments, one or more blocks of FIG. 6 may be implemented by one or more other machines not depicted in the figure.

[0090] The method 600 may include receiving a selection of at least one of a first manufacturing process or a first manufacturing facility for performing manufacturing operations of the first manufacturing process, and identifying modifications and / or optimizations to improve eco-efficiency (reducing resource consumption and / or environmental impact). The manufacturing process may be associated with a manufacturing system (e.g., manufacturing system 102 of FIG. 1 ).

[0091] At block 601, processing logic receives a first selection of at least one of a first manufacturing process or a first manufacturing equipment (e.g., tool 202 and subcomponent 204 of FIG. 2) for performing manufacturing operations of the first manufacturing process.

[0092] At block 602, processing logic inputs the first selection into a digital replica of the first manufacturing facility (e.g., digital replica 400 of FIG. 4). In some embodiments, the digital replica may include a physics-based model (e.g., physics-based model 214 of FIG. 2).

[0093] At block 603, processing logic determines environmental resource usage data indicative of a first environmental resource consumption and / or environmental impact of a first manufacturing process performed on the first manufacturing equipment based on the physical conditions of the first manufacturing process. In some embodiments, the environmental resource usage data includes at least one of energy consumption, gas consumption, or water consumption associated with the first manufacturing equipment for performing manufacturing operations of the first manufacturing process.

[0094] In some embodiments, the processing logic uses the environmental resource usage data as input to a machine learning model. The processing logic further includes obtaining one or more outputs of the machine learning model, the one or more outputs being indicative of the modification. In some embodiments, the one or more outputs of the machine learning model may further indicate a confidence that the modification, when implemented, will reduce the first environmental resource consumption and / or environmental impact of the manufacturing process. The processing logic further determines that the confidence in the modification satisfies a threshold condition.

[0095] In some embodiments, the first environmental resource consumption and / or environmental impact includes environmental resource consumption and / or environmental impact associated with one of a replacement procedure or a maintenance procedure for a consumable part of the first manufacturing facility.

[0096] In some embodiments, the environmental resource usage data includes life cycle data for components associated with the manufacturing facility. The remediation may further include implementing corrective actions associated with the components.

[0097] At block 604, processing logic determines modifications to the first manufacturing process that reduce environmental resource consumption and / or environmental impact (e.g., resource consumption per unit) of the first manufacturing process performed on the first manufacturing facility. In some embodiments, the modifications include changing one or more of the manufacturing process steps and / or manufacturing facility processing parameters. In some embodiments, prioritize optimization of one or more of energy consumption, gas consumption, or water consumption of the first manufacturing facility for performing manufacturing operations of the first manufacturing process.

[0098] At block 605, processing logic optionally applies the corrections to the manufacturing process. At block 606, processing logic optionally provides the corrections for display by a graphical user interface (GUI). In some embodiments, multiple corrections are determined and provided for presentation on the graphical user interface (GUI) (e.g., to a user, such as a system operator). In some embodiments, the corrections are presented to the user in ranked order by confidence. In some embodiments, the corrections are presented to the user with a visual indicator representing the confidence associated with each normative behavior. For example, one or more corrections having the highest confidence may be depicted in a first color (e.g., green or gold), and one or more corrections having a confidence near a threshold level may be depicted in a second color (e.g., yellow or silver). In some embodiments, the corrections may be placed in hierarchies or groups based on their associated confidence.

[0099] In some embodiments, the processing logic may further determine, based on the environmental resource usage data, that the first selection does not satisfy a threshold eco-efficiency. The processing logic may further perform optimization of the first selection in response to determining that the first selection does not satisfy the threshold eco-efficiency. Performing optimization of the first selection includes ascertaining one or more modifications to the first selection that, when applied, result in an updated eco-efficiency that satisfies the threshold eco-efficiency.

[0100] In some embodiments, the processing logic is further to receive a second selection of manufacturing equipment for performing a second manufacturing operation in the second manufacturing process. The processing may receive second sensor data associated with the second manufacturing operation from one or more sensors associated with the second selection of manufacturing equipment. The processing logic may further update the digital replica to generate an updated digital replica, the updated digital replica being associated with the first and second selections of manufacturing equipment. The processing logic may further obtain one or more outputs from the digital replica including aggregated environmental resource data indicative of environmental resource consumption and / or environmental impact (e.g., per unit consumption and / or impact) of the first and second selections.

[0101] 7A-C are flow diagrams of methods 700A-C associated with identifying modifications to a manufacturing process according to some embodiments of the present disclosure. Methods 700A-C may be performed by processing logic, which may include hardware (e.g., circuitry, dedicated logic, programmable logic, microcode, a processing device, etc.), software (e.g., instructions running on a processing device, a general-purpose computer system, or a dedicated machine), firmware, microcode, or a combination thereof. In some embodiments, method 700A may be performed in part by machine learning system 170 (e.g., server machine 172, dataset generator 174, etc.). Machine learning system 170 may use method 700A to at least one of train, validate, or test a machine learning model according to embodiments of the present disclosure. In some embodiments, one or more operations of method 700A may be performed by dataset generator 174 of server machine 172. In some embodiments, methods 700B-C may be performed in part by machine learning system 170 (e.g., server machine 172, server machine 180, server machine 192, etc.). Machine learning system 170 may use method 700B to train a machine learning model according to an embodiment of the present disclosure. Machine learning system 170 may use method 700C to use a trained machine learning model according to an embodiment of the present disclosure. In some embodiments, one or more operations of methods 700B-C may be performed by correction identification component 194 of machine learning system 170. Note that components described with respect to one or more of FIGS. 1-6 may be used to illustrate aspects of FIGS. 7A-C. In some embodiments, a non-transitory storage medium stores instructions that, when executed by a processing device (e.g., of machine learning system 170), cause the processing device to perform methods 700A-C.

[0102] For simplicity of explanation, methods 700A-C are depicted and described as a series of acts. However, acts in accordance with the present disclosure may occur in various orders, simultaneously, in parallel with multiple instances per store, and / or with other acts not shown and described herein. Furthermore, not all of the illustrated acts may be performed to implement methods 700A-C in accordance with the disclosed subject matter. Additionally, those skilled in the art will understand and appreciate that methods 700A-C may alternatively be represented as a series of interrelated states via a state diagram or events.

[0103] Referring to FIG. 7A, a method 700A is associated with generating a dataset for a machine learning model to process a selection of manufacturing processes and / or manufacturing equipment to identify modifications to inputs.

[0104] At block 702, processing logic implementing method 700A initializes a training set T to an empty set.

[0105] At block 704, processing logic generates a first data input (e.g., a first training input, a first validation input) that includes a selection of a manufacturing process and manufacturing equipment.

[0106] In some embodiments, at block 706, processing logic generates a first target output for one or more of the data inputs (e.g., the first data input). The first target output may be, for example, a modification to the manufacturing process and / or manufacturing equipment. The processing logic may generate the target output based on an input selection of the manufacturing process and / or manufacturing equipment.

[0107] At block 708, processing logic optionally generates mapping data indicating an input / output mapping. The input / output mapping (or mapping data) may refer to data inputs (e.g., one or more of the data inputs described herein), target outputs for the data inputs (e.g., the target outputs identify output data), and associations between the data inputs and the target outputs. The processing logic may, for example, perform gradient descent and backpropagation to update weights for nodes in one or more layers of the machine learning model.

[0108] At block 710, processing logic adds the data input generated at block 704 and / or the mapping data generated at block 708 to the dataset T.

[0109] At block 712, processing logic branches based on whether dataset T is sufficient for at least one of training, validating, and / or testing machine learning model 190. If so, execution proceeds to block 714; otherwise, execution continues back to block 704. In some embodiments, the sufficiency of dataset T may be determined simply based on the number of input / output mappings in the dataset, while in some other embodiments, the sufficiency of dataset T may be determined based on one or more other criteria (e.g., a measure of diversity of the data examples, accuracy, etc.) in addition to or instead of the number of input / output mappings.

[0110] At block 714, processing logic provides dataset T (e.g., to server machine 180) for training, validating, and / or testing machine learning model 190. In some embodiments, dataset T is a training set and is provided to training engine 182 of server machine 180 to perform training. In some embodiments, dataset T is a validation set and is provided to validation engine 184 of server machine 180 to perform validation. In some embodiments, dataset T is a test set and is provided to test engine 186 of server machine 180 to perform testing. In the case of a neural network, for example, input values ​​(e.g., numerical values ​​associated with data inputs) of a given input / output mapping are input to the neural network, and output values ​​(e.g., numerical values ​​associated with target outputs) of the input / output mapping are stored at output nodes of the neural network. Connection weights in the neural network are then adjusted according to a learning algorithm (e.g., backpropagation, etc.), and this procedure is repeated for other input / output mappings in dataset T. After block 714, the machine learning model (e.g., machine learning model 190) may be at least one of trained using training engine 182 of server machine 180, validated using validation engine 184 of server machine 180, or tested using testing engine 186 of server machine 180. The trained machine learning model may be implemented by correction identification component 194 (of machine learning system 170) to generate output data for further use by server 120 procedures (e.g., component synthesis tool 122, digital replica tool 124, optimization tool 126, recipe builder tool 128, and / or resource consumption tool 130).

[0111] Referring to FIG. 7B, a method 700B is associated with training a machine learning model to determine modifications to manufacturing processes and / or manufacturing equipment that improve eco-efficiency.

[0112] At block 720, processing logic receives a selection of a manufacturing process or manufacturing facility.

[0113] In some embodiments, processing logic identifies a label corresponding to a modification to the manufacturing process at block 722. In some embodiments, the label indicates the modification to the manufacturing equipment part and / or the manufacturing process and an associated reduction in environmental resource consumption and / or environmental impact.

[0114] At block 724, processing logic trains a machine learning model using data inputs including manufacturing process data (e.g., and target outputs including labels) to generate a trained machine learning model configured to generate outputs (e.g., modifications) that can be applied to the manufacturing process to reduce environmental resource consumption and / or environmental impact.

[0115] In some embodiments, the machine learning model is trained using unsupervised learning based on the data input (e.g., without a target output) to generate a trained machine learning model (e.g., to cluster the data). In some embodiments, the machine learning model is trained using supervised learning based on the data input and the target output to generate a trained machine learning model.

[0116] Referring to FIG. 7C, method 700C is associated with using machine learning models to determine modifications to manufacturing processes and / or manufacturing equipment to improve (e.g., optimize) eco-efficiency (e.g., reduce environmental resource consumption and / or environmental impact).

[0117] At block 740, processing logic receives current manufacturing process data. At block 742, processing logic provides the current data (e.g., manufacturing process data) to a trained machine learning model. The trained machine learning model may be trained according to method 700B.

[0118] At block 744, processing logic obtains one or more outputs from the trained machine learning model. In some embodiments, the outputs include modifications to the manufacturing process and / or manufacturing equipment that, when implemented, improve the environmental efficiency of the manufacturing process and / or manufacturing equipment. At block 746, processing logic causes application of one of more manufacturing process modifications to the manufacturing process based on the outputs.

[0119] 8 depicts a block diagram of an example computing device operating in accordance with one or more aspects of the present disclosure. In various illustrative examples, various components of computing device 800 may represent various components of client device 150, server 120, data store 112, and machine learning system 170 illustrated in FIG.

[0120] The exemplary computing device 800 may be connected to other computer devices within a LAN, an intranet, an extranet, and / or the Internet (e.g., using a cloud environment, cloud technology, and / or edge computing). The computing device 800 may operate in the capacity of a server within a client-server network environment. The computing device 800 may be a personal computer (PC), a set-top box (STB), a server, a network router, switch, or bridge, or any device capable of executing (sequentially or separately) a set of instructions that specify actions to be taken by the device. Furthermore, while only a single exemplary computing device is illustrated, the term "computer" shall also be deemed to include any collection of computers that individually or together execute a set (or sets) of instructions to perform any one or more of the methodologies discussed herein.

[0121] The example computing device 800 may include a processing device 802 (also referred to as a processor or CPU), a main memory 804 (e.g., read-only memory (ROM), flash memory, dynamic random access memory (DRAM) such as synchronous DRAM (SDRAM), etc.), a static memory 806 (e.g., flash memory, static random access memory (SRAM), etc.), and a secondary memory (e.g., a data storage device 818), which may communicate with each other via a bus 830.

[0122] The processing device 802 represents one or more general-purpose processing devices, such as a microprocessor, a central processing unit, and the like. More specifically, the processing device 802 may be a complex instruction set computing (CISC) microprocessor, a reduced instruction set computing (RISC) microprocessor, a very long instruction word (VLIW) microprocessor, a processor implementing other instruction sets, or a processor implementing a combination of instruction sets. The processing device 802 may also be one or more special-purpose processing devices, such as an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA), a digital signal processor (DSP), a network processor, or the like. According to one or more aspects of the present disclosure, the processing device 802 may be configured to execute instructions implementing the methods 600, 700A-C illustrated in FIGS. 6-7.

[0123] The example computing device 800 may further include a network interface device 808 that may be communicatively coupled to a network 820. The example computing device 800 may further include a video display 810 (e.g., a liquid crystal display (LCD), touch screen, or cathode ray tube (CRT)), an alphanumeric input device 812 (e.g., a keyboard), a cursor control device 814 (e.g., a mouse), and an audio signal generating device 816 (e.g., a speaker).

[0124] The data storage device 818 may include a machine-readable storage medium (or, more specifically, a non-transitory machine-readable storage medium) 828 on which one or more sets of executable instructions 822 are stored. According to one or more aspects of the present disclosure, the executable instructions 822 may include executable instructions associated with performing the methods 600, 700A-C illustrated in FIGS.

[0125] The executable instructions 822 may also reside, completely or at least partially, within the main memory 804 and / or within the processing device 802 during execution thereof by the exemplary computing device 800, with the main memory 804 and the processing device 802 also constituting computer-readable storage media. The executable instructions 822 may further be transmitted or received over a network via the network interface device 808.

[0126] Although computer-readable storage medium 828 is shown as a single medium in FIG. 8, the term "computer-readable storage medium" shall be considered to include a single medium or multiple media (e.g., centralized or distributed databases, and / or associated caches and servers) that store one or more sets of operating instructions. The term "computer-readable storage medium" shall also be considered to include any medium that can store or encode a set of instructions for execution by a machine that cause the machine to perform any one or more of the methodologies described herein. The term "computer-readable storage medium" shall therefore be considered to include, but not be limited to, solid-state memory, and optical and magnetic media.

[0127] Some portions of the above detailed descriptions are presented in terms of algorithms and symbolic representations of operations on data bits within a computer memory. These algorithmic descriptions and representations are the means used by those skilled in the data processing arts to most effectively convey the substance of their work to others skilled in the art. An algorithm is here, and generally, conceived to be a self-consistent sequence of steps leading to a desired result. The steps require physical manipulations of physical quantities. Usually, though not necessarily, these quantities take the form of electrical or magnetic signals capable of being stored, transferred, combined, compared, and otherwise manipulated. It has proven convenient at times, principally for reasons of common usage, to refer to these signals as bits, values, elements, symbols, characters, terms, numbers, or the like.

[0128] It should be borne in mind, however, that all of these and similar terms are to be associated with the appropriate physical quantities and are merely convenient labels applied to these quantities. Unless specifically stated otherwise, as will be apparent from the discussion that follows, throughout the description, discussions utilizing terms such as "locate," "determine," "store," "adjust," "cause," "return," "compare," "create," "stop," "read," "copy," "throw," "exchange," "execute," or the like, will be understood to refer to the acts and processes of a computer system or similar electronic computing device that manipulate and convert data represented as physical (electronic) quantities in the computer system's registers and memory into other data similarly represented as physical quantities in the computer system's memory or registers or other such information storage, transmission, or display device.

[0129] Examples of the present disclosure also relate to an apparatus for performing the methods described herein. This apparatus may be specially constructed for the required purposes, or it may be a general-purpose computer system selectively programmed by a computer program stored on the computer system. Such a computer program may be stored on a computer-readable storage medium, such as any type of disk, including, but not limited to, optical disks, compact disk read-only memories (CD-ROMs), and magneto-optical disks, read-only memories (ROMs), random access memories (RAMs), erasable programmable read-only memories (EPROMs), electrically erasable programmable read-only memories (EEPROMs), magnetic disk storage media, optical storage media, flash memory devices, other types of machine-accessible storage media, or any type of media suitable for storing electronic instructions, each coupled to a computer system bus.

[0130] The methods and displays presented herein are not inherently related to any particular computer or other apparatus. Various general-purpose systems can be used with programs in accordance with the teachings herein, or it may prove convenient to construct more specialized apparatus to perform the required method steps. The required structure for a variety of these systems appears to be set forth in the description below. Additionally, the scope of the present disclosure is not limited to any particular programming language. It will be understood that a variety of programming languages ​​can be used to implement the teachings of the present disclosure.

[0131] It should be understood that the above description is illustrative and not restrictive. Many other embodiments will be apparent to those skilled in the art upon reading and understanding the above description. While the present disclosure describes particular examples, it should be recognized that the systems and methods of the present disclosure are not limited to the examples described herein, but may be practiced with modification within the scope of the appended claims. Accordingly, the specification and drawings are to be regarded in an illustrative rather than a restrictive sense. The scope of the present disclosure should, therefore, be determined with reference to the appended claims, along with the full scope of equivalents to which such claims are entitled.

Claims

1. receiving, by a processing device, a first selection of at least one of a first manufacturing process or a first manufacturing facility for performing manufacturing operations of the first manufacturing process; inputting the first selection into a digital replica of the first manufacturing facility, the digital replica outputting physical conditions of the first manufacturing process; determining environmental resource usage data indicative of a first environmental resource consumption of the first manufacturing process performed on the first manufacturing facility based on the physical conditions of the first manufacturing process; determining, by the processing device, modifications to the first manufacturing process performed on the first manufacturing facility that reduce the environmental resource consumption of the first manufacturing process; at least one of a) applying the modification to the first manufacturing process; or b) providing the modification for display by a graphical user interface (GUI); A method comprising:

2. Determining the modification comprises: using the environmental resource usage data as an input to a machine learning model; obtaining one or more outputs of the machine learning model; and wherein the one or more outputs indicate the modification.

3. The one or more outputs further indicate a confidence that the modification, if implemented, will reduce the first environmental resource consumption of the manufacturing process, and the method further comprises: The method of claim 2 , further comprising determining that the confidence in the modification satisfies a threshold condition.

4. determining, based on the environmental resource usage data, that the first selection does not satisfy a threshold eco-efficiency; performing an optimization of the first selection in response to determining that the first selection does not meet the threshold eco-efficiency; and and wherein performing the optimization of the first selection includes ascertaining one or more modifications to the first selection that, when applied, result in an updated eco-efficiency that satisfies the threshold eco-efficiency.

5. determining a process parameter value window for a process parameter associated with a first one of the manufacturing operations; determining an environmentally optimized process parameter window within the process parameter value window; 10. The method of claim 1, further comprising: wherein process parameter values ​​within the environmentally optimized process parameter window cause the first manufacturing operation to consume a reduced amount of resources compared to process parameter values ​​outside the environmentally optimized process parameter window.

6. 2. The method of claim 1, wherein the environmental resource usage data includes at least one of energy consumption, gas consumption, or water consumption associated with the first manufacturing facility for performing the manufacturing operations of the first manufacturing process.

7. 7. The method of claim 6, wherein the modification is associated with a prioritized optimization of one or more of the energy consumption, the gas consumption, or the water consumption of the first manufacturing facility for performing manufacturing operations of the first manufacturing process.

8. The method of claim 1 , wherein the digital replica comprises a physics-based model of the first manufacturing facility for performing the manufacturing operations of the first manufacturing process.

9. receiving a second selection of manufacturing equipment for performing a second manufacturing operation in a second manufacturing process; receiving second sensor data associated with the second manufacturing operation from one or more sensors associated with a second selection of the manufacturing equipment; updating the digital replica to generate an updated digital replica, the updated digital replica associated with the first and second selections of manufacturing facilities; obtaining one or more outputs from the digital replica including aggregate environmental resource data indicative of environmental resource consumption of the first selection and the second selection; The method of claim 1 further comprising:

10. The method of claim 1 , wherein the first environmental resource consumption comprises environmental resource consumption associated with one of a replacement procedure or a maintenance procedure for a consumable part of the first manufacturing facility.

11. The method of claim 1 , wherein the environmental resource usage data includes life cycle data for a component associated with the manufacturing facility, and the modification includes performing a corrective action associated with the component.

12. 1. A method for training a machine learning model to identify modifications to a manufacturing process or to a selection of manufacturing equipment for performing manufacturing operations of said manufacturing process, said modifications being associated with reducing environmental resource consumption of said manufacturing process, said method comprising: generating, by a computing device, training data for the machine learning model, wherein generating the training data includes: identifying a first training input having a first selection of a first manufacturing process; identifying a first target output for the first training input, the first target output comprising a first modification to the first manufacturing process that, when applied, reduces a first environmental resource consumption of the first manufacturing process; providing, by the computing device, the training data to train the machine learning model on (i) a set of training inputs including the first training inputs and (ii) a set of target outputs including the first target outputs, wherein the trained machine learning model will receive as input a new selection of a new manufacturing process and will produce new outputs based on the new inputs, the new outputs indicating new modifications to the new manufacturing process that, when applied, reduce new environmental resource consumption of the new manufacturing process; A method comprising:

13. 13. The method of claim 12, wherein generating the training data further comprises inputting the first selection into a digital replica of manufacturing equipment associated with the first manufacturing process, the digital replica outputting physical conditions of the first manufacturing process.

14. 13. The method of claim 12, wherein the new environmental resource consumption comprises one of energy consumption, gas consumption, or water consumption associated with new manufacturing equipment that performs manufacturing operations of the new manufacturing process.

15. The method of claim 12 , wherein each training input in the set of training inputs is mapped to a target output in the set of target outputs.

16. Memory and a processing device coupled to the memory; a processing device for processing a signal from the processing device; receiving a first selection of at least one of a first manufacturing process or a first manufacturing facility for performing manufacturing operations of the first manufacturing process; inputting the first selection into a digital replica of the first manufacturing facility, the digital replica outputting physical conditions of the first manufacturing process; determining environmental resource usage data indicative of a first environmental resource consumption of the first manufacturing process performed on the first manufacturing facility based on the physical conditions of the first manufacturing process; determining modifications to the first manufacturing process performed on the first manufacturing facility that reduce the environmental resource consumption of the first manufacturing process; a) applying the modifications to the first manufacturing process; or b) providing the modifications for display by a graphical user interface (GUI).

17. The processing device further comprises: using the environmental resource usage data as an input to a machine learning model; obtaining one or more outputs of the machine learning model; and wherein the one or more outputs indicate the modification.

18. The processing device further comprises: determining, based on the environmental resource usage data, that the first selection does not satisfy a threshold eco-efficiency; performing an optimization of the first selection in response to determining that the first selection does not meet the threshold eco-efficiency; and and to perform the optimization of the first selection, the processing device is to ascertain one or more modifications to the first selection that, when applied, result in an updated eco-efficiency that satisfies the threshold eco-efficiency.

19. 17. The system of claim 16, wherein the environmental resource usage data includes one of energy consumption, gas consumption, or water consumption associated with the first manufacturing facility for performing manufacturing operations of the first manufacturing process.

20. 17. The system of claim 16, wherein the digital replica comprises a physics-based model of the first manufacturing facility for performing the manufacturing operations of the first manufacturing process.

Images