Download PDF

Biomanufacturing, a process in which organisms and their biological systems are used to produce chemicals and biomaterials, has been a part of the military industrial base since World War I. The fermentation- based Weizmann process produced acetone to address the solvent shortage that had curtailed the production of cordite, a smokeless gunpowder critical to the Allied war effort in the early 20th century.¹ In its infancy, biomanufacturing relied on the availability of naturally occurring microbial strains that produced specific materials. Resulting products were therefore confined to the natural genetic armamentarium of a few easily grown strains and relied on brute-force mutagenesis and screening techniques to optimize production.

While the repertoire of known natural products is quite large, the search for an industrially tractable microorganism for any given product remained both time-consuming and labor-intensive. Additionally, the number of available biological processes used for materials production continued to remain relatively small in contrast to the diverse methodologies employed by their petrochemical industry competitors.

The limited repertoire of the industry in the 20th century pales in comparison to today’s biotechnology industry, in which advancements in the fields of synthetic biology, artificial intelligence, and robotics have resulted in the rapid expansion of small-scale production capabilities and the development of corresponding high-throughput experimentation platforms that accommodate both generation and downstream processing of biologically derived materials and are poised to revolutionize the way civilian and military sectors produce materials.²

Modern biomanufacturing combines a variety of disciplines, including engineering, biology, chemistry, and computer science, and facilitates the production of biologically derived materials on a commercial scale. Much like operations in a traditional distillery, the operations of a biomanufacturing plant can be separated into two phases—fermentation and product recovery. In the fermentation phase, sugars and other nutrients are converted into biomass and the desired product (for example, starch into yeast and alcohol). In the product recovery phase, the alcohol is distilled and purified.

The technical advances of the past 30 years and the newfound cognizance of supply-chain vulnerabilities arising from the COVID-19 pandemic and the Russian invasion of Ukraine have set the stage for biomanufacturing’s reemergence as a means of domestic chemical and material production, while the versatility of modern bioengineering is pushing the boundaries of traditional materials science. Consequently, biotechnology can make a significant contribution to addressing vulnerabilities in Department of Defense (DOD) supply chains.³ Advances in biotechnology continue to provide innovative tools and capabilities to serve both DOD missions and commercial needs, particularly for unique materials whose manufacture requires advanced or unconventional methods of production. Supported by initiatives such as the Defense Advanced Research Projects Agency’s Living Foundries program,⁴ modern biotechnology continues to advance the manufacture of bioderived materials through techniques such as enzyme engineering, cell-free reactions, and the expansion of classic fermentation-based technologies to previously unexploited host microbes.⁵ Accordingly, initiatives from both the White House and DOD now promote the establishment of domestic supply chains that use biotechnology- based materials and biochemicals for high-value chemical precursors, military armor, energetics, and propellants.⁶

Recognizing its potential to revitalize supply chains, DOD recently named biotechnology as one of its modernization priorities.⁷ As a result, BioMADE, a new manufacturing innovation institute, was formed to shepherd products from bench to commercialization. In addition, several internal research efforts have been announced, some examples of which can be found in table 1.

 

Military-focused biotechnology research and development efforts, however, have remained largely investigator-driven, both in sponsored academic research programs and at DOD internal Service laboratories. Consequently, transitions occur largely through ad hoc connections, with few clear, formalized mechanisms for initiating advanced stages of development or accession of the products as technology insertions into programs of record. Modernization efforts are beginning to address this bottom-up approach, a route that causes substantial technical risk during the transition into commercial-scale production.⁸ Indeed, retrospective analysis of several well-known early commercial failures in the industrial biotechnology sector identified communication breakdowns between teams responsible for different phases of the development cycle as a common root cause of failure and extended transition delays.⁹ Fortunately, DOD is well-versed in supporting iterative project communication in the context of its established acquisitions project management framework, and the industrial biotechnology and traditional acquisitions pathways share some parallels that will inform future DOD biomanufacturing efforts.

Here, we propose a model for biotechnology development for military products that leverages insights from both DOD’s traditional acquisition modes and meta-analysis of decades of industrial biotechnology expertise. The model includes building integrated product teams (IPTs) that span the development cycle and align with core principles of successful full-scale biomanufacturing, such as:

• “beginning with the end in mind”

• designing early-stage experiments with constraints of the final production facility in mind, or “scaling down to scale up”

• using project management principles founded in systems engineering

• prioritizing secure, data-driven communications.

We believe that through this model innovative breakthroughs at the discovery level can more reliably be carried through to a full-rate manufacturing capability and delivered more rapidly across the proverbial “valley of death” into the hands of the warfighter.

The Biomanufacturing Development Cycle

The development cycle of any given biotechnology product (see figure) comprises four major phases—benchscale strain and process development, scale-up and de-risking, process integration, and full-scale production. These phases have analogues in the traditional DOD acquisitions pathway (see below). Biomanufacturing project development typically begins as proof-of-principle experiments that demonstrate the ability to produce a given bioproduct in a small-scale, benchtop environment. Genes encoding new biochemical pathways are introduced into the organism’s life-coding DNA and through a series of optimization steps that channel metabolites—internal intermediate chemicals—into the product and away from nonproductive pathways, a strain is developed and optimized to produce the compound or material of interest. The design-build-test-learn phase is analogous to spiral development in software acquisitions and is in many cases implemented on automated, high-throughout experimental platforms. As development advances, progressively larger scales of production are employed in pilot scale manufacturing facilities, at first with physically unlinked fermentation and recovery operations, culminating in the construction and commissioning of the final production facility.

Because the goal of the biotechnology development cycle is to produce commercial or military-relevant quantities of product in a large-scale facility, it is critical to design the early-phase experiments to mimic as closely as possible the production environment. Early-stage experiments often use the “best of all possible worlds” scenarios, including well-oxygenated, complex media and optimal gene expression conditions. Furthermore, product extraction and purification methods can be executed at small scales without incurring significant costs, particularly those associated with waste handling. Utilities and energy costs are also typically not considered at that stage.

In contrast, the large scale of fermentation facilities used for biomanufacturing operations imposes physical constraints on biological processes, such as the ability to deliver sufficient oxygen or maintain a “perfectly mixed” growth environment.¹⁰ The realities of construction, operation, maintenance, utilities, and waste disposal impose further limitations, any one of which could render a given strain, process, or product economically or technically nonviable. Accordingly, transition of small-scale biotechnology successes to larger scales can falter, potentially with substantial financial impacts.¹¹ In some cases, one may be able to address such delays through iterative rounds of smaller scale testing to determine performance parameters more closely based on the real-world operating conditions. Alternatively, in a worst-case scenario, an entirely new microorganism, metabolic pathway, or fermentation strategy might be required.¹² Consideration of technical and economic constraints at the beginning of project development can substantially reduce the average piloting timeline (currently 6 months to 3 years, on average) and thereby shorten the time to market. Ideally, integration of the strain, process, and manufacturing development cycles would facilitate completion of detailed design and initiation of construction of a full-scale industrial facility within 18 months of piloting a given process.¹³

Building and operating large biomanufacturing facilities involve notable financial risk, with construction costs for full-scale facilities ranging around $100 million or more. For the time being, industrial biotechnology will continue to scale using process-scaling methods rather than the model-based methods routinely employed in the chemical industry, owing to the need to use live organisms as catalysts. Using organisms as catalysts means model simulations would be required to run not just 5 or 10 reactions to simulate experimental conditions, but rather many thousands of reactions in parallel. Perturbations in the growth conditions, even seemingly small ones, can dramatically impact yield, and therefore it has been seen during industrial scaling processes that the largest fixed infrastructure envisioned imposes an operational upper limit to all process variables that should also be considered during earlier small-scale efforts.¹⁴

While maximum stirring speed and oxygen availability are often cited as the most critical constraints, temperature, pressure, and supplemental capabilities such as sensor feedback requirements can also be limiting contributors.¹⁵ Encouragingly, recognition of these constraints by the field has increasingly resulted in industrial and academic community adoption of the scale-downto- scale-up principle, which is particularly discussed in the context of constraints on oxygen-transfer rates and mechanical agitation and mixing since these typically do not scale easily from the benchtop level to larger manufacturing-scale volumes and operations.¹⁶ Still, as process development schemes are often considered lucrative trade secrets, a lack of standardization in industrial process requirements continues to mean that ad hoc development outside of an IPT structure can be an approach fraught with risk.

Integrating Bioindustry “Best Practices” With DOD Acquisition Principles

Thus, the state of the art for process- scaled biomanufacturing technologies means that, at least initially, development of promising DOD biochemical and material projects would benefit from bringing together basic and applied biotechnology researchers with seasoned process-development and manufacturing teams into an IPT. This would allow DOD researchers the opportunity to learn and apply industrial transition strategies for new materials and products while reducing technological risk profiles for industrial scaling partners. To facilitate this modification to the bottom-up transition of DOD biotechnology products, we suggest a framework for future biotechnology product development efforts that is modeled on elements of traditional DOD acquisition programs.

The later development cycle of a biotechnology product shares some features with the canonical acquisition cycle for major defense acquisition programs (MDAPs), such as major weapons systems. While not sharing the high program costs of an MDAP, the progression from discovery to full-rate manufacturing naturally reflects the systems engineering strategy that is the foundation of larger acquisition programs, offering some insights into how DOD biotechnology development efforts might be structured. DOD defines systems engineering as “the seamless and efficient integration of both new and existing technologies so that the finished product is greater than the sum of its component parts.”¹⁷ In an MDAP, five major phases of the system development and deployment occur:

• materials solution analysis (MSA)

• technology maturation and risk reduction (TMRR)

• engineering manufacturing and development

• production and deployment

• operations and support.

The first four of these have analogous phases to the development of biomanufactured products. The phases are separated by milestones, at which point decisions can be made to proceed to the next phase. It is important to note that the conventional technology readiness level or manufacturing readiness level (MRL) metrics are not easily applied to biotechnology products; rather, a BioMRL scheme has been proposed to accommodate the unique properties of biological processes.¹⁸

Scaling from discovery to manufacturing of a biotechnological product follows a similar trajectory. Proof-of-principle experiments both in fermentation and product recovery occur during the MSA phase (pre–Milestone A), initiating a design-build-test-learn cycle that culminates with strains and separation processes that can transition to manufacturing (Milestone B). In industry, the same teams are involved in the pre– and post–BioTech Milestone A phases since those teams are the most knowledgeable regarding the needs of the project. During the iterations, robust discussion between the demonstration and manufacturing teams can contribute to a more efficient development process for robust performance at larger volumes. This type of interactive collaboration across the traditional Milestone gateways forms the backbone of the biotech industry’s IPTs.

Iterative cross-Milestone performance development also is not entirely new to the acquisition process; it is also found in DOD’s software acquisitions pipeline.¹⁹ Parallels between microbial strain development and software program development have often been drawn in the industrial space. In the bioindustrial model, specified capital infrastructure— the “hardware”—consists of fermentation tanks, downstream chemical processing equipment, and supporting utilities that have defined physical limits of performance. Meanwhile, microorganisms producing biochemicals and biomaterials of interest—the “software”—are developed to perform optimally within the specifications set by the available capital infrastructure.²⁰ Both development cycles rely on all participants working with the same understanding of operational requirements and available infrastructure. Similarly, both require clear communication between development and implementation teams to efficiently launch optimizations and improvements during rollout and often, once operational, represent a key reason for use of program management approaches based on systems engineering.

Recommendations

A review of the bioindustrial literature reveals several “lessons learned” during industrial scale-up projects. Incorporating these insights can offer guiding principles for DOD research and development (R&D) initiatives to increase the likelihood of successful transition of novel fermentation and bio-based processes to full-scale operation.

Foremost, top-down communication of manufacturing constraints and an emphasis on beginning with the end in mind allow for the use of strategies that feasibly translate from small-scale bench explorations to commercial implementation.²¹ For example, conditions found in 10,000-liter bioreactors can now be mimicked in small, parallelized 0.25- liter reactors. Applying the above software development approach to defining process constraints allows for iterative development of biocatalysts until the process reaches an optimized state, where “the [hardware] dictates how good the [software] has to be.”²²

Additionally, early efforts in assembling an IPT to establish understanding of the logistical operations involved in transitioning from a bench-scale to a fullscale operation can ensure that all essential data is collected during the project development campaign. Robust and strategic process development mitigates increased risks incurred when adopting new technologies previously unproved at scale. It can also be advantageous to “be the second mouse” where relevant—essentially, allowing other first adopters to encounter and troubleshoot the technology and then adopting or adapting the technology once the first bugs have been addressed.²³ Finally, a techno-economic analysis feasibility engineering evaluation approach, developed using iterative inputs from both techno-economic analysis (TEA) and operability insights, is often used to provide a framework for targeted project development as early as capability conceptualization. Together these industrial lessons learned can aid DOD researchers, performers, and program managers in structuring biotechnology programs.

Engineer the Biology With the Final Production System and Economics in Mind. Companies that tout IPT-based R&D approaches to develop bioproducts in-house from strain engineering through to commercialization often implement elements of feasibility engineering evaluation (FEE) methodology to rapidly transition between scaled-down and scaled-up systems. Similar to the systems engineering approach used by the National Aeronautics and Space Administration (NASA)²⁴ for developing software packages, FEE methodology considers the project from the vantage point of the full-scale working system and then employs periodic assessments to evaluate the extent to which overall performance goals are being met, in some instances requiring developmental iteration to achieve targeted performance metrics.²⁵ The assessment benchmarks are typically defined for both economic and performance insights. Economic data includes the costs of material, equipment, utilities, facilities, operators, land, and asset costs. Performance insights are based on titer, rate, and yield (TRY) metrics.

In the DOD space, process constraints are typically further defined with respect to military specification requirements for equivalent “drop-in” materials, including down to the level of color, type, and concentrations of specific contaminants. Definitions of the stages of FEE can vary slightly, but they are often categorized as FE-1, “preliminary process design”; FE-2, “process design package”; and FE-3, “basic engineering package,” leading to detailed engineering and construction.²⁶ Table 2 details data package components often included in each FEE stage. Revisions to performance expectations, and potential need for additional TRY development, might occur at any FEE stage and should especially be evaluated once the purchase-ready major equipment specifications are established in stage FE-3, if the project is using commercial off-the-shelf units in a software/ hardware development approach.

The techno-economic analysis performs a critical function in the design of strains, processes, and production facilities. The TEA is a comprehensive cost- and resource-level analysis of all inputs and outputs of a proposed process leading to an assessment of commercial viability. TEAs are increasingly developed early in applied biotechnology development, establishing quantifiable metrics to assess viability and risk or compare different process option.²⁷ One use for a TEA is to assess the piloting data insights as a theoretical full-scale model diagram using current economic data, specifically including the defined process-flow diagram that identifies all material, energy, labor, capital, and operational expenditures. Projections are based on established engineering heuristics and, often, hard-earned experience. Depending on the sophistication of the model, TEAs can be used to assess the effects of changes to process designs, resource and disposal costs, labor, and economic factors on the overall cost of the product and approximate the price point at which a material will be profitable or an investment can be recovered.

Finally, a TEA should be iterative and become more granular as the information about a process becomes more refined and the project moves through the FEE stages. For example, the Advanced Research Projects Agency–Energy Reducing Emissions using Methanotrophic Organisms for Transportation Energy program conducted a preliminary TEA that detailed the key metrics of a biochemical production process and analyzed economic sensitivities to changing feedstock and utilities prices for relevant scenarios as early as the first round of applied research funding, which in DOD is at the 6.2 level.²⁸ There should be some flexibility for defining major TEA metrics for each project of interest, but minimum thresholds for technology transfer during process scaling are already established in industry. While somewhat variable based on the intended use of the biochemical or biomaterial and whether alternative routes to its production exist, some rules of thumb specify go/no-go transition thresholds as:

• a 30-percent cash cost advantage over the best-available commercial technology for established chemicals

• 3- to 10-fold improvement in properties over existing commercial alternatives for novel materials²⁹

• replacement or augmentation of an unreliable or foreign source for a critical material.

 

To transition bioproducts created within DOD, we propose first assessing scale-up feasibility using a preliminary TEA (pre-TEA), which occurs during the materials solution analysis phase prior to Milestone A. This assessment should ideally include a contextual understanding of the envisioned market space (or operational needs) and an understanding of what is required to reach market and/or military standards for the product. These standards should form the benchmarks for every checkpoint throughout the remainder of the project. Benchmarks for performance would be somewhat variable for each product, but a framework would likely include:

• Strain or bioprocess performance: Are iterative development cycles approaching TRY metrics that support commercial viability? One g/liter is an often-cited benchmark for fermentation, but nonfermentation bioprocesses will vary by product. What is the iterative rate of improvement? Are the strains genetically stable?

• Product recovery: Do established downstream processing (DSP) methods exist? Are they robust? How much additional development is needed? What is the percentage of recovery of the target fermentation products? Seventy percent overall recovery is a commonly seen benchmark, but it will vary by product and DSP method. How does the observed recovery impact the TRY metrics? Is overall process performance (total yield, percentage of product recovered, purity) predictable?

• Military applications: What is the impact of this product on military systems? If the product is novel, what might be the impact on near-peer products? What research is needed to confirm performance? What volume of material is required to demonstrate this performance? Is there a civilian use for this product, and will it impact the market for this product?

At the second benchmark, just prior to Milestone B (see figure), more detailed process development information should be available and can be used to narrow the uncertainty windows. Data collected prior to the second assessment should also be used to produce an initial process flow diagram as the entry into FEE stage FE-1 of the engineering and manufacturing development portion of an MDAP, the review of which might already reveal areas for further iterative development at smaller scales prior to Milestone B or even Milestone A.

While ultimately reaching industrial go/no-go standards is desirable, in some cases the warfighter advantage might take precedent over economics. Additional or alternative criteria could apply to military- critical materials, such as remediating a single point of failure within a supply chain, establishing a domestic source for a critical material, or even developing a novel material with no near-peer comparison. In these cases, factors may favor advancement of a process when cost factors or properties are neutral or even disadvantageous.

Collaboratively Develop Strain Improvement and Downstream Processing Strategies With Scale in Mind for Strategic DOD Biomaterial Development. To ensure that early-stage development efforts are matched to endstate production requirements, biotechnology development efforts should adopt program management frameworks that incorporate expertise drawn from the entirety of the development cycle but can pivot in the face of changing technology and end-user expectations. Diane DiEuliis and coauthors recently pointed out the mismatch of the broader system-focused DOD acquisition process with the nimble, small-company bioeconomy sector.³⁰ In their commentary, the authors made several wide-ranging recommendations, including advocating for programs that explore “new governance and engagement concepts over the life cycle of several exemplar products, from discovery to fielding.” The IPT is a multidisciplinary group of individuals who together assume responsibility for delivering a product or process.³¹ IPTs are mandatory for large DOD acquisition programs and provide a potential model management of the development of military biotech products from concept through fielding.

Indeed, recent DOD-wide efforts have begun to address the issue of integrating biotechnology development processes. Notably, the Office of the Secretary of Defense’s Synthetic Biology for Military Environments (SBME) effort under the Applied Research for the Advancement of Science and Technology Priorities program began to bridge the gaps between the various Service labs, forming interdisciplinary teams across multiple Service laboratories to apply synthetic biology tools to address military-relevant problems, such as biosensors to detect toxic metals in drinking water and modulating microbiota to enhance warfighter performance.³² But while the SBME achieved success in building such interdisciplinary teams between the Service labs, this program did not include strategies to move its successful products into production, and First proof-of-concept project for Army Combat Capabilities Development Command Chemical Biological Center biomanufacturing facility will be production of fuel for Hellfire missile; hence transitions from the SBME have relied on ad hoc efforts.

A different approach is taken by the Biotech Optimized for Operational Solutions and Tactics program,³³ which aims to bridge the “valley of death” for targeted materials between basic/early applied research and development to manufacturing. Unlike previous efforts, this program requires that performers identify a strategy to transition at the proposal stage and makes that a significant component of the evaluation of proposed work. Both programs, however, assume a linear progression from early- phase discovery through scale-up and transition, without the iterative feedback that has facilitated successful industrial scale-up efforts. To maximize the impact of biotechnology investments, achieve successful transitions into working products, and accelerate the process of scaling production from lab to plant, we recommended that DOD biotechnology researchers adopt key industry best practices that have led to market adoption of commercial products at scale.

In manufacturing, the process makes the product. Companies that focus on using fermentation to make products optimize both process development and strain performance. These companies emphasize the importance of “early and frequent” communication between teams because reaching a biomaterial product of sufficient quality and quantity requires a cross-disciplinary approach. They note that the biomaterial production process—fermentation—is only the beginning of making a bioproduct and that close collaboration with downstream recovery results in significant savings in time and costs.³⁴ Rather than waiting until a material has reached a specific threshold to initiate product recovery concept efforts, bringing in the DSP team during the earliest bioproduct idea development phase could not only help set scale-up feasibility targets that assist the fermentation team in setting targets for technology transition but also sometimes influence the final form of product that the strain team chooses to pursue.³⁵ TRY metrics are major design variables for DSP development.³⁶ Moreover, meeting overall project goal benchmarks requires collaboration from both the strain and DSP teams because both product and byproduct yields substantially impact DSP efficiency and thereby the TEA. Ultimately, the performance of the team’s approach to the entire process—with respect to both quality and quantity—determines the efficiency and thereby the cost-effectiveness of the full-scale operation.

When a bioproduct is identified for development into military use, an IPT would coalesce with representation from three major phases of the development cycle:

• Manufacturing group: Representing the “ultimate customer,” a team of primarily engineers and scientists building the full-scale facility defines and communicates requirements for process development constraints with bench and pilot scale teams responsible for strain and process development as early as the point of project conceptualization. Possible constraints on process development include economic, physical, environmental, and technical limitations. In the DOD ecosystem, this group would include the members of the BioMADE Manufacturing Innovation Institute and other industrial transition partners. It should also include eventual end-users of precursor biomaterial products that require further processing who can contextualize process development operations, such as the program executive offices responsible for requirements development, later stage testing and evaluation, and integration into weapons systems.

• Process development group: Responsible for scale-up performance and evaluation, a team of engineers and scientists that bridges the manufacturing, strain development, and DSP chemistry teams (see below). This team is responsible for translating, modeling, and developing scale-up processes that fit within the operational bounds defined by the manufacturing group in a process that is often referred to as “de-risking.”³⁷ These efforts produce both techno-economic analyses as well as process operability performance insights. During scale-up, the process development team identifies and mitigates process bottleneck points, additional safety considerations, and process failure points. In the DOD ecosystem, this group includes the U.S. Army Combat Capabilities Development Command Chemical Biological Center’s BioManufacturing Facility and/or the BioMADE Manufacturing Innovation Institute facility at the University of Minnesota, which was slated for construction in 2023.³⁸ DOD-focused pilot facilities could also play a role in the TMRR phase that is not typically served by commercial contract manufacturing facilities by “seeding the market” or producing enough of the desired material to enable researchers at other DOD labs to fabricate test articles and prototype items using the new materials. Seeding the market is particularly valuable when developing new materials for which end-uses are still being identified. In addition, DOD-focused pilot facilities can play key roles in the engineering manufacturing and development phase by bridging the economic and technical “valley of death” between lab development and industrial adoption and by de-risking the manufacturing process for new DOD applications.

• Strain development and DSP chemistry teams: Responding to calls for new biomaterials R&D, a team primarily of scientists creates and optimizes microbial strains or bio-based processes to make essential and new biochemical and biomaterial products with yields that meet the key performance parameters defined by the manufacturing and process development groups. The strain teams develop microbial strains specifically for industrially viable processes, while the DSP chemistry teams develop scalable biomaterial recovery processes with final products that meet performance requirements. In the DOD ecosystem, this group includes performers from the Army Research Laboratory, Army Research Office, Air Force Research Laboratory, Naval Research Laboratory, and National Reconnaissance Office. Regular communication between strain and separations teams is essential to develop strains and processes that meet key performance parameters.³⁹ For example, rather than developing a strain that produces the exact molecule being targeted, the manufacturing and process-development groups may request that a strain be developed that produces a derivative of the target molecule that facilitates a less energy-, materials-, or waste-intensive separation process. The derivative could then be converted to the desired product later in the process using conventional chemical or enzymatic approaches.

Since both a high-yield strain or bio-based process and a productive separation strategy are critical to success of a biotechnology product, biotechnology development programs place high value on concurrent and collaborative benchscale development of both aspects. Most fundamental, such development should occur at the early stages, in line with recommendations of established biochemical production companies, that have adopted the mantra, We do not do experiments in the plant—experimentation happens at bench and pilot scale.⁴⁰ Ultimately, early implementation of an IPT structure ensures that insights capable of making process development successful are rapidly shared across the multidisciplinary team.

Prioritize Infrastructures for Efficient Communication and Sharing of Data. The multidisciplinary nature of military biomanufacturing IPTs will inevitably bring vocabulary-based challenges. Within a biomaterials scale-up pipeline, contributing disciplines may include microbiology, synthetic biology, bioengineering, chemistry, materials science, chemical engineering, process engineering, industrial engineering, mechanical engineering, and materials engineering—each with its own technical arcana and jargon. However, numbers, math, and related data analysis represent a unifying language that cuts across all science, technology, engineering, and mathematics fields. Methods for evaluating project performance can include benchmarks, thresholds, set point values, rates, yields, concentrations, efficiencies, titers, and percent conversion, among others. Emphasizing cross-discipline and data-driven communication, such as that which ideally occurs within the structure of an MDAP, could put each piece of the work in context and highlight concrete contributions by teams toward the ultimate achievement of project goals. Clear data-driven communication between disciplines during biomaterial scale-up is essential not only for optimum performance but also for feasibility, safety, and logistical discussions.

In the growing DOD biomanufacturing ecosystem, increased emphasis on discussion of translational principles among manufacturing, pilot, and strain and separations development groups has the potential to substantially assist the efficiency of biomaterial scale-up efforts. To achieve this goal, the different IPT components will need to “speak a common language,” or else have a translation strategy, to ensure that technical communications occur in the proper mathematical formats, with the appropriate science and engineering units. A spectacular and infamous example of a communication failure can be drawn from NASA’s Mars Climate Orbiter mission, in which a misunderstanding of the units used for critical calculations between the U.S. and European Union teams sent the orbiter too close to the planet and into its atmosphere, where it burned up.⁴¹ An average biomaterial scale-up project should expect to span units from micrograms to kilograms and in some cases tons. Emphasizing review of mathematical translations among the bench, pilot, and full-scale operations will be essential to accurate data analysis foundations.

Effective data-driven communication also requires strategic collection of relevant design data at each stage of the scale-up process. Within the biomanufacturing sector, a few pioneers have embraced technology-driven scale-up strategies. While highly efficient robotic workflows and high-throughput data collection have transformed strain development at the bench level, at the pilot and manufacturing scales capital and operational realities limit multi-replicate data collection operations. Accordingly, with successive increases in scale, strategic data capture and analysis opportunities become increasingly discrete and infrequent. At the same time, real-time data collection driven by sensors and process control software becomes increasingly automated with increasing scale, and greater data granularity can drive machine-learning insights toward the long-term development of smarter biomanufacturing process control strategies.⁴²

Scale-up of biomanufacturing processes bridges basic research funded or executed by DOD to the broader manufacturing ecosystem and the supply chain. Industrial biotechnology’s answers to address these operational boundaries come from decades of hard-earned insight. Just as DOD R&D already implements and uses expert practices for the development of biotechnological chassis organisms at the lab scale, the tried-andtrue techno-economic analysis feasibility engineering evaluation approach for scale-up in industrial biotechnology could serve as a model for development of a robust scale-up strategy for DODrelevant biotechnology products.⁴³ Such a strategy emphasizes ongoing and pervasive technical communication among basic and applied research labs, industrial partners, and government stakeholders, with the goal of implementing industrial design standards from the time the decision to produce a given product is made. Shifting focus from the idea of “transitioning up to check a box” to the concept of “designing with the end in mind” would align DOD biomanufacturing modernization initiatives with established industry practices and traditional acquisition frameworks, allow for the standardization of biotechnology development, facilitate the adoption of an iterative “software development” acquisitions mentality for biotech projects, and increase return on R&D investment dollars while shortening the time required to achieve field-ready performance. JFQ