ISPE 专家博客探讨自动化跨洁净级别物料传递的技术进展与行业举措（第二部分）

However, the technological landscape is evolving. New generations of cleanroom-compatible autonomous mobile robots, automated decontamination concepts, and industry-wide collaborative initiatives are beginning to address the key technical barriers.

This second part explores these developments in more detail. It highlights recent advances in cleanroom robotics, discusses emerging automated decontamination techniques, and reviews ongoing industry efforts aimed at enabling fully automated cross-grade material transfer.

Recent technological advancements

Recent technological advancements have introduced mobile robots that are cleanroom-compatible up to Grade D or C—and in some cases, even theoretically suitable for Grade B or A environments. These mobile robots are engineered to endure standard decontamination procedures in GMP settings, including disinfectant wiping and vaporized hydrogen peroxide (VHP) treatment. Their designs support cleanroom airflow requirements and ensure full cleanability, including the underside of the vehicle.

A notable development in this field is the Stäubli Sterimove™, a cleanroom-compatible mobile robot currently being tested by Novo Nordisk in various applications within Grade D and C environments1. While the Sterimove™ is engineered with features aligned with higher-grade use, practical deployment in higher grades environments has not yet been validated, and passing through airlocks is explicitly out of scope in the current evaluation. The platform offers key features such as application-specific top modules, full VHP compatibility via complete encapsulation, and a design that allows for cleaning from underneath.

While the base module is supplied by Stäubli, it is important to note that the application-specific top module must be developed and provided by a specialized systems integrator. The complete system—base and top module—–must comply with all applicable GMP requirements for its intended use.

Among the applications currently under evaluation is environmental monitoring using the contact plate method. In this setup, the top module incorporates a cleanroom-compatible robotic arm that manipulates contact plates, establishes controlled contact with designated surfaces, and stores the plates appropriately for later incubation and analysis.

Other applications focus on material handling tasks within cleanroom environments, where the Sterimove™ transports items within controlled zones in compliance with cleanroom operational standards.

Integrating mobile robots into GMP cleanroom environments necessitates careful quality assessment, particularly of the cleaning procedures and operational processes associated with their use. Any implementation must be qualifiable. The overall system—including robot movement patterns, interaction with cleanroom airflow, and impact on operations—must comply with GMP standards. Robot trajectories may need to be restricted to validated paths.

While many mobile robot surfaces are designed to be cleanable, wheels remain a critical area. Critical aspects include particle generation by friction and the risk of spreading contamination. Procedures for cleaning and disinfection of the wheels must be assessed in terms of efficacy. These could include manual wiping or rolling over sticky mats.

The path towards automated material transfer

While cleanroom-compatible mobile robots represent significant progress in automating material movement within classified environments, the broader challenge of enabling fully automated material transfer between different cleanroom grades remains unresolved. This next section explores the emerging concepts, pilot initiatives, and enabling technologies that aim to close this gap. From industry-led proof-of-concept efforts to advancements in automated decontamination, UV disinfection, and robotic unpacking, these developments collectively represent the foundational steps toward realizing end-to-end automated material transfer in GMP-regulated settings.

The BioPhorum proof-of-concept workstream

Within the BioPhorum Operations Group (BPOG), the Technology Strategy Phorum has launched a Proof of Concept (PoC) workstream focused on the topic of automated material transfer2. This initiative aims to explore and validate opportunities for automation that eliminate the need for human intervention, enabling reliable, low-intervention material movement.

The workstream specifically targets the transfer of materials from Controlled Non-Classified (CNC) spaces to Grade D cleanroom environments without manual operator-driven decontamination. Demonstrating this capability would represent a significant advancement, enabling broader deployment of mobile robots in biomanufacturing environments. Ultimately, this work supports the biopharmaceutical industry’s long-term vision of highly automated, low-intervention manufacturing—driving efficiency, improving compliance, and reducing operational risk.

As one of the first major deliverables, the group has developed a user requirement specification (URS), which defines the functional, system, and safety requirements for an automated material transfer solution between CNC and Grade D cleanroom environments.

The URS outlines the prerequisites, process flow, and design needs for implementing mobile robots and automated airlocks. It covers both greenfield and brownfield facility scenarios and emphasizes minimizing human intervention to reduce contamination risk, improve efficiency, and enhance data integrity.

While the URS provides a comprehensive blueprint for automating material transfer from CNC to Grade D environments, it also acknowledges certain limitations. Notably, some proposed technologies—such as disinfection using UV—are not yet recognized by regulatory authorities, requiring further validation and industry collaboration. Additionally, the URS focuses on conceptual and functional requirements, and does not provide detailed engineering specifications, leaving site-specific adaptations and risk assessments to individual facilities.

As a next phase of the workstream, a dedicated, pre-competitive industry initiative has been launched to establish the efficacy of UV as a disinfection technology3. This effort brings together biomanufacturers and technology providers to develop standardized validation frameworks and generate shared data on feasibility and efficacy. Addressing the current lack of validated data and regulatory guidance is critical to enabling risk-based acceptance of UV in GMP environments. Through benchmarking, protocol development, data sharing, and regulatory engagement, the initiative aims to support broader adoption of UV-based disinfection—unlocking automation of material transfers, reducing reliance on manual, chemical-based cleaning processes, and contributing to more efficient and sustainable operations.

Automated decontamination

To achieve fully automated material transfer across cleanroom grades, the decontamination process within the material airlock remains the central technical and regulatory hurdle. While recent advancements in automated material handling systems now support operations within specific cleanroom grades, crossing between zones with differing GMP classifications still requires effective, validated decontamination—a function that, to date, relies heavily on manual procedures.

VHP

VHP disinfection is a proven and widely accepted method for microbial decontamination during material transfer in pharmaceutical environments. Its effectiveness is well established, particularly in systems designed for transferring single-use tubs of components into sterile areas like fill lines. As with all other technologies mentioned in this section, VHP alone also cannot effectively cover both cleaning and disinfection—see Table 1.

Table 1: Comparison of cleaning and disinfection methods in terms of efficacy

However, VHP has notable limitations—most importantly, its long cycle time. Before the cycle, a VHP unit needs pre-cleaning of the chamber and the materials that are to be transferred. A full disinfection cycle, including degassing, can take up to 40 minutes before the opposite side of the airlock can be safely opened. The time highly depends on the size and load of the chamber. This delay limits throughput and makes VHP less suitable for processes requiring high-frequency or continuous material transfers. Instead, VHP is better suited for batch-wise, high-volume transfers where the extended cycle time can be amortized across larger loads.

Material compatibility is another constraint. VHP residuals and permeability issues can limit what can safely be passed through the airlock, restricting flexibility in handling diverse components. Furthermore, these systems are typically designed for batch transfer and are size-constrained, making them less ideal for modern manufacturing models that emphasize continuous flow and modularity.

Hydrogen peroxide-based material airlocks (MALs) have been used for decades, but their application is associated with stringent facility requirements and relatively long decontamination and aeration cycles due to the chemical’s toxic properties. Implementing such systems in existing (brownfield) facilities would typically require a retrofit of both the MAL unit itself and the HVAC system. While newer approaches using atomized hydrogen peroxide are emerging, their implementation in MAL systems remains to be demonstrated.

While VHP remains a robust option for specific use cases, its suitability must be re-evaluated in the context of evolving production needs and the growing push toward continuous, automated material handling.

UV

Automated decontamination involves two critical components: particulate removal and microbial reduction. For microbial decontamination, ultraviolet (UV) disinfection is gaining attention due to its non-contact application, scalability, and potential for integration into automated workflows. UV only poses a hazard during exposure, and exposure occurs only while the UV system is running, with no residual harm—unlike VHP, which requires a “40-minute” downtime before material retrieval. Once the UV cycle is completed, the material may be removed immediately.

However, successful implementation of UV disinfection as an automated technique requires careful definition of key parameters, including:

• Distance from the light source to the object
• Irradiation angle and beam uniformity
• Exposure time, tailored to microorganism resistance and material surface
• Geometry of the payload, which can introduce shadow zones
• Particles removal

Ensuring comprehensive surface coverage may require the strategic placement of multiple UV sources, including beneath the material, to mitigate the risk of untreated areas.

In parallel to cleanroom-mobile robot PoC initiatives, a separate team at Novo Nordisk is conducting feasibility studies focused on automating the decontamination step. These include evaluations of both air showers and UV-based disinfection systems. The team is currently engaged in early-stage validation in collaboration with quality control and regulatory functions, aiming to demonstrate efficacy, repeatability, and compliance with GMP standards. Formal integration of this work into the broader BioPhorum PoC initiative is under active consideration.

Air shower

For reduction of particulate contamination, air showers have emerged as the most promising automated approach. They are efficient for gross cleaning in a non-classified or CNC area. Adhering contamination (sticky particles) are not removed by an air shower, only loose particles (dust). Authority acceptance is generally limited to non-classified and CNC applications.

Wiping robot

While automated systems like air showers can effectively remove non-adhering particulates such as dust, they fall short when dealing with gross or adhering contamination—such as worst-case scenarios involving strongly adhering residues (e.g. dried product splashes, sticky disinfectant residues, or grease-like contamination transferred via secondary packaging). This limitation is particularly relevant for unclassified (UC) to CNC transfers, where a spray-and-wipe cleaning concept is commonly applied. In these cases, mechanical action (i.e., wiping) remains essential. Traditionally, this is performed manually by operators after a spray-down with disinfectant. Automating this step poses significant challenges. A conceptual analogy might be an automated car wash, where rotating brushes adjust to the shape of a vehicle. In pharmaceutical material transfers, the variability in box and pallet shapes is somewhat more controlled—typically cuboidal—making generalization possible. However, key limitations arise: cleaning materials like wipes are typically preferred to be single-use (especially in higher grades) to minimize cross-contamination risk, although reusable wiping tools could be acceptable in UC to CNC applications provided appropriate cleaning, segregation, and validation concepts are in place, ruling out conventional rotating brush systems in many GMP-relevant scenarios due to hygiene requirements.

A more promising avenue involves robotic wiping, where a robotic arm equipped with perception tools (e.g. 3D vision and distance sensors) could mimic manual cleaning paths using disposable wipes. While technically feasible, this approach is still in early development stages, and current technologies are not yet mature enough for routine use in cleanroom environments.

Table 2: Comparison of cleaning and disinfection methods - pros and cons

Automated unpacking

In addition to particulate and microbial decontamination, a third contamination control strategy is multi-layer wrapping, where materials are sealed in multiple layers. During transitions between cleanroom grades the outer layer is removed as part of the airlock procedure, particularly for items like consumables and gowning. Materials to which this approach is applicable must still be included in the materials transfer validation. The manner of unwrapping plays a critical role—for instance, if an operator touches only the outer surface with disinfected gloves and pushes the inner material out without direct contact, it may be acceptable without wiping.

Automating this unpacking step presents a significant technical challenge. It requires advanced robotic capabilities such as object recognition, tactile sensing, and precise motion control. Soft, deformable, or transparent packaging—like plastic bags—further complicates robotic perception and manipulation. Tasks such as tearing seals, cutting bags, or peeling off outer layers are still beyond the capabilities of commercially available systems.

So-called no-touch systems are a solution already on the market for introducing materials (such as pre-sterilized primary packaging material) into filling lines by the means of tubs. For lower grades and diverse material formats, production-ready solutions are yet to be developed.

While AI-driven solutions, including large behavior models and training via simulation or telemanipulation, show promise, these approaches are still in early development stages and not yet production-ready. In parallel, adapting packaging designs to be more robot-friendly—by including visual markers or standardized gripping points—may offer a complementary path forward.

Solving this problem will require close collaboration between experts in computer vision, cognitive robotics, automation design, and packaging engineering.

Next steps

Advancing automated decontamination across cleanroom transitions requires a clear, focused approach, beginning with the selection of a specific transition to target. The movement of materials from CNC to Grade D presents the most accessible starting point, as it involves relatively fewer regulatory and technical barriers. However, the Grade D to C transition may offer greater long-term value, since many critical decontamination steps—such as the manual removal of visible and adhering contamination—are already well established between CNC and D.

To address this challenge effectively, pharmaceutical companies should adopt a step-by-step strategy that identifies key technical and quality hurdles while engaging cross-functional stakeholders across Quality, CQV, Engineering, and Manufacturing. Cleaning procedures associated with each cleanroom grade and airlock type must be systematically assessed and benchmarked against emerging automated solutions to identify feasibility gaps and opportunities for improvement.

Given that these enabling technologies—such as automated decontamination, UV disinfection, and robotic unpacking—are not directly tied to proprietary drug formulations or production processes, industry-wide collaboration is both possible and essential. By sharing learnings, aligning requirements, and pooling validation efforts, companies can collectively accelerate the development of solutions that are not only technologically advanced, but also GMP-compliant and regulatory-ready.

Ultimately, joint investment and data generation across the industry will be key to establishing the credibility of automated decontamination solutions. A coordinated approach to validation and regulatory engagement can pave the way for broader acceptance and implementation—unlocking the full potential of automation in pharmaceutical cleanroom material transfer.

Conclusion

Automating material transfer across cleanroom grades is not a single technological challenge—it is a systems problem that spans robotics, sterility assurance, contamination control, packaging design, and regulatory compliance.

While significant progress has been made in cleanroom-compatible robotics and automated intralogistics, the automation of the decontamination step within material airlocks remains the critical barrier. Technologies such as UV disinfection, automated air showers, robotic wiping systems, and robotic unpacking offer promising directions, but further development, validation, and regulatory alignment are required before they can be deployed at scale.

Achieving this vision will require cross-industry collaboration. Because automated material transfer technologies are not tied to proprietary drug products, pharmaceutical companies have a unique opportunity to share learnings, align requirements, and collectively generate the validation data needed to support regulatory acceptance.

With continued innovation and collaboration, the industry can move toward a future where automated, low-intervention material transfer becomes a foundational element of highly automated pharmaceutical manufacturing.