You are here

Resilient Chemical Manufacturing


OUSD (R&E) MODERNIZATION PRIORITY: Artificial Intelligence (AI)/Machine Learning, Autonomy TECHNOLOGY AREA(S): Biomedical, Chemical/Biological Defense, Information Systems, Materials/Processes OBJECTIVE: Resilient Chemical Manufacturing (RCM) seeks to enable the rapid reallocation and optimization of existing domestic chemical manufacturing infrastructure to a new suite of products, allowing the U.S. to leverage existing onshore production equipment to respond to chemical supply chain disruptions. DESCRIPTION: The United States relies on chemical manufacturing to provide products ranging from everyday consumer goods (plastics, fabrics, adhesives, paints) to cutting edge technologies (medicines, electronic materials), industrial goods (dyes, pesticides) and military supplies (fuels, explosives). While many high-volume, petroleum-derived chemical feedstocks are produced domestically, much of the fine chemical manufacturing necessary for complex chemical products (e.g., pharmaceuticals, electronics, energetics) has been outsourced. As a result, the U.S. is vulnerable to dynamic factors that are challenging to forecast, including issues as complex as political conflict, as unpredictable as natural disasters, and as simple as economies of scale. While the origin might vary, the impact is universal – such forces disrupt our supply of chemical feedstocks and products, affecting critical sectors of our nation including defense, healthcare, transportation, communications, and the economy. While developing new manufacturing infrastructure and methods (e.g., automated, distributed, continuous) is one way to address these challenges, another approach of specific interest to DARPA is to build software planning capabilities that enable automated allocation and optimization of existing domestic chemical manufacturing infrastructure to a new set of products. Conventional plant-based chemical manufacturing consists of diverse sets of equipment (reactors, pumps, columns, separators, etc.) connected in a defined sequence to produce a single product. Allocation and reconfiguring of this equipment to produce a different product is a slow, manual operation, requiring detailed process knowledge and deep expertise on a given product. As a result, domestic manufacturing capacity for any new product is vastly underestimated, and diverse, secondary considerations related to critical manufacturing process attributes (e.g., scale, purity, and throughput; geographic location/distribution; and site-specific regulatory considerations) are challenging to consider and impossible to fully optimize. Developing the capacity to automatically identify, allocate, and optimize chemical manufacturing assets across multiple sites/vendors and understand dependencies of particular assets on user requirements for new chemical products would revolutionize our ability to address chemical supply chain challenges across multiple sectors. RCM will enable rapid reallocation of existing domestic chemical manufacturing to produce chemicals that are subject to supply chain disruptions, allowing the U.S. to leverage on-shore, U.S.-owned production equipment to meet demand for chemicals due to supply chain disruptions or other dynamic demand swings. RCM will build robust production planning algorithms for a variety of domestic and foreign chemical products critical to the U.S. industrial and consumer base, develop precise ontologies for manufacturing equipment, establish a dynamic database of U.S.-owned manufacturing assets, and demonstrate a software tool that can pair production needs with latent (yet-to-be-configured) manufacturing capacity. Importantly, RCM will not develop new production infrastructure, but instead provide the capacity to model and forecast existing production equipment to meet a new production need. PHASE I: This topic solicits Direct to Phase II proposals ONLY. Proposers must demonstrate that the following has been achieved outside of the SBIR program: Initial software tool/prototype that is capable of automated allocation of domestic manufacturing assets for at least ten chemical products. The demonstrated capability must include: (1) a database of domestic chemical manufacturing assets, (2) an ontology to adequately describe and measure equivalency of chemical manufacturing equipment, (3) the ability to consider process features (e.g., volume, chemical compatibility, temperature ranges) and user requirements (e.g., throughput, purity, regulatory standards), and (4) capacity to consider equipment and/or processes across multiple manufacturing sites. PHASE II: RCM performers will build and validate software that enables automated allocation, management, and optimization of domestic chemical manufacturing assets. DARPA anticipates approaches that include (1) acquisition of domestic manufacturing asset information resulting in a dynamic asset database; (2) economic, security, and availability assessments of existing critical fine chemicals with approaches to computationally assess substitute chemicals; (3) fully operational, validated software with a user interface (UI) designed for non-experts that automatically allocates domestic chemical manufacturing assets across the U.S. to a particular chemical in shortage; and (4) a suite of tools that enables optimization across both chemical feedstock and/or supply chain availability and domestic manufacturing potential. Base Period (24 Months): Phase II fixed payable milestones for this program should include: • Month 1: Report on current asset database and plans to incorporate additional elements, to include key details such as equipment, specifications, manufacturing locations, chemicals, suppliers, quantities, country of origin, etc., as required, to support technology/software development milestones and deliverables throughout the effort. The report should highlight current database knowledge gaps and a plan to acquire additional information to expand the breadth, scope, and utility of the database. • Month 3: Report on selection of at least 10 chemicals that represent critical precursors, fine chemicals, and/or feedstocks to important chemical products, along with synthetic routes relevant to proposed efforts that will serve as a testbed for demonstration and validation of technology deliverables over the course of the award. Selected molecules should be directly applicable to at least one critical supply category (e.g., semiconductors and critical electronic components, energetic materials, active pharmaceutical ingredients (API)) as outlined in the 2021 House Armed Services Committee Report of the Defense Critical Supply Chain Task Force1. Selection of final testbed molecules will be approved after consultation with DARPA. • Month 5: Report on initial algorithm development, software architecture, and modeling approaches, along with potential operational/user features of the software prototype to be employed for the Month 9 demonstration. The Month 5 report should also include details of security controls relative to database content and access that ensures vendor proprietary information is protected. • Month 9: Report summarizing Month 9 software prototype demonstration. The report should provide details on approach, prototype architectures and algorithms, data sets, and results demonstrating initial proof-of-concept performance of software prototype (without experimental/manufacturing validation) to identify alternative/re-purposed manufacturing infrastructure or substitute chemical feedstocks/precursors. The Month 9 demonstration must utilize two of the 10 selected testbed molecules under three variable manufacturing/supply-chain scenarios selected by DARPA. The report should also detail software performance relative to database composition (e.g., number of vendors, types of equipment, etc.) with a plan to expand, augment, and refine database content and quality to enhance software/algorithm performance and capabilities. • Month 12: Report on lessons learned, updated architectures, algorithms, and learning approaches based on results/analysis of software prototype performance during Month 9 demonstration to include critical aspects of information contained in the database as well as a plan for experimental validation of asset allocation by Month 21. • Month 15: Report describing expansion and optimization of technology platform integrating production capacity, logistics, costs, sustainability, and stakeholder constraints relevant to the proposed efforts. • Month 18: Report describing the development of advanced tools and features that simplifies software operation (e.g., user interface and operability) and improves performance (e.g., time to provide a result, additional feature selection including process features and/or user requirements). The report should also include details related to development of the user interface, search, command, and control functions enabling use by non-experts. • Month 21: Report on (1) initial software design and engineering for graphical user interface; visual analytics; and search, command, and control to include details/findings of beta-testing activities with non-experts and (2) details of experimental validation runs to include validation of user-defined requirements/inputs (e.g., throughput, purity, etc.) from the software realized in a chemical manufacturing facility. • Month 24: Final demonstration and report documenting version 2.0 prototype architectures and algorithms, methods, results, and performance of software platform to identify alternative/re-purposed manufacturing infrastructure or substitute chemical feedstocks/precursors specific to three additional testbed molecules under five variable manufacturing/supply-chain scenarios selected by DARPA. The report should also detail software performance relative to usability by non-experts and to key data/metrics contained in the database with a plan to expand, augment, and refine database content and quality to enhance software/algorithm performance and capabilities if needed. Option 1 (12 Months): • Month 28: Report on development and performance of optimized user interface, cyber security features, cloud infrastructure, and/or software package intended for deployment and commercialization. Report should document subcontractors and vendors along with strategies for product launch, production, marketing, sales, and technical support, as appropriate. • Month 34: Capstone demonstration to stakeholders as defined in consultation with DARPA. • Month 36: Final report documenting software prototype architectures and algorithms, methods, results, and performance of software platform to identify alternative/re-purposed manufacturing infrastructure or substitute chemical feedstocks/precursors specific to the remaining five testbed molecules under seven variable manufacturing/supply-chain scenarios selected by DARPA. In addition, the final report should include quantitative metrics on decision making benefits, costs, risks, and schedule for implementation of a full prototype capability based on the pilot demonstrations. This report shall include an identification of estimated level of effort to integrate the pilot capability into an operational environment, addressing computing infrastructure and environment, decision making processes, real-time and archival data sources, and maintenance and updating needs; reliability, sensitivity, and uncertainty quantification; and transferability to other military users and problems. The report shall also document any scientific advances that have been achieved under the program. (A brief statement of claims supplemented by publication material will meet this requirement), and final PI meeting presentation material. PHASE III DUAL USE APPLICATIONS: Fine chemical precursors are essential to a wide variety of applications critical to national security and defense such as plastics, adhesives, energetics, electronic materials, and pharmaceuticals. As such, RCM has broad applicability within the DoD, the broader U.S. Government, and the commercial sector to include other manufacturing sectors as well as supply chain management. REFERENCES: 1. U.S. House Armed Services Committee: Defense Critical Supply Chain Task Force Final Report (2021) coral Leptastrea purpurea. Scientific Reports, 2019. 9(1): p. 2291. KEYWORDS: Model-based systems engineering, fine chemical manufacturing, logistics and supply chain, domestic manufacturing infrastructure, automated asset allocation, information technology, AI algorithms, materials databases, modeling and simulation, active pharmaceutical ingredients, energetic materials, Agile manufacturing, Computer-aided process planning, Decision theory, Distributed manufacturing, Manufacturing inventory systems, Logistics systems, Model-based quality control, Predictive modeling, Process diagnosis, Process planning, Production optimization, System simulation, Statistical process control
US Flag An Official Website of the United States Government