Scientists and lab managers must keep pace with the demands of new therapy development — often with reduced budgets. Combined with competing pressures like inflation, supply chain challenges, staffing issues, and the need to operate sustainably, the environment makes it essential that lab managers have the insights to drive both efficiency and discovery.
How can lab managers continually balance the need to invest time in the core scientific tasks that drive research with non-core activities, like inventory management? Artificial intelligence (AI) offers one path forward. These systems, supported by machine learning (ML) and automation technology, are transforming the lab ecosystem.
Turning actionable data into streamlined success
Historically, labs used a just-in-time approach to determine inventory levels, ordering product and consumables so that they arrived as needed. In the aftermath of post-COVID global supply chain challenges, however, lab managers have been more likely to maintain higher stock levels, sometimes bulk-buying to ensure supply. While this just-in-case approach can provide an inventory cushion if the supply chain falters, the excess inventory also ties up money and valuable lab space.
With AI, lab managers can operate on a hybrid of just-in-time and just-in-case. AI considers a wide range of inputs from throughout the workflow, analyzing data points provided by electronic lab notebooks, smart shelves, lab applications and systems, etc. The technology can then look at historical and current consumption along with other factors to generate prescriptive analytics.
For example, an AI system might determine that a specific consumable will run out of stock in five days. AI can provide further insights by determining that the vendor’s shipping time is exceeding five days, so the product won’t arrive at the lab in time. That predictive insight allows the lab manager to decide if they want to wait for the product, request an accelerated shipping time or consider an alternate solution. The result is more efficient and more sustainable decision-making that is proactive instead of reactive.
An AI system can also recognize inventory anomalies. It’s not unusual for a lab to see spikes in usage of specific products or consumables. When lab or procurement managers view these spikes as a new trend, it can skew forecasting and cause excess ordering. However, AI can identify spikes as anomalies and factor that into its forecast, creating a more accurate assessment that helps reduce or eliminate overstock.
Lab Quality Management Certificate
The Lab Quality Management certificate is more than training—it’s a professional advantage.
Gain critical skills and IACET-approved CEUs that make a measurable difference.
AI is a journey
Like adopting any new process in the lab, successful AI implementation doesn’t happen overnight — in fact, you must crawl before you can walk or run. These best practices can guide your AI journey:
- Collect as much data as possible with core applications. Data is currency for AI. Digitalizing non-research operations, like management of inventory, chemicals, and equipment, helps create a foundational database from which AI can operate. What’s more, as the number of data points increase across the lab ecosystem, so does the accuracy of AI’s output. Tracking lab assets as much as possible gives lab managers — and the AI system — the transparency needed for successful AI implementation.
Examples of core application tools include:
Smart buttons: Lab personnel use these small devices to request on-demand services, like quality control (QC), glass wash service, or safety emergency assistance. Placed throughout the lab workspace, smart buttons boost efficiency by serving as a single point of contact that eliminates the need for a researcher to leave the bench. With one click, a researcher can request an item or medical waste removal. They’re particularly helpful in clean room environments because they reduce the need for additional human presence in that space.
Machine learning can then use smart button data to help lab managers answer the question: How is my lab operating? For example, the data provides insights into what types of assistance, products or services team members request, and when they request them. This perspective, when combined with other data flows, allows lab managers to take a proactive approach to streamlining processes.
Smart shelves: This technology is equipped with digital sensors that detect inventory changes at the piece-level. The information is transmitted to inventory management software, which provides real-time visibility into inventory and consumables stock. Pre-set guardrails trigger auto-replenishment or stocking as needed.
With these insights, lab teams can better understand consumption to identify fast- and slow-moving inventory items. It also provides the real-time data that AI combines with historical data to prescribe order and replenishment cycles. Smart shelves eliminate the need for personnel to inspect and scan shelved inventory, allowing researchers to spend more time on science.
AI-powered vision systems: Computer vision can accurately assess inventory levels, which AI can then use to determine if restock or reorder is needed. One advantage of this vision technology is that it can make sense of even chaotic stockrooms by determining what inventory is on a shelf even if the individual stock pieces are out of position or jumbled together.
As this technology develops, AI-powered vision systems could be positioned on the bench to assist scientists. For example, a vision system could alert a scientist that a specific asset is missing from the workspace or that there is a tripping hazard in the lab.
- Automate and connect as many data interactions as possible. Many end users in the lab only work in the application that’s relevant to them, restricting their visibility across the full workflow. A scientist may have little real-time access to understand if a consumable is available or if equipment is properly calibrated and ready to use. As a result, connecting the data points in an easy-to-access, personalized dashboard is an essential step in the AI journey.
Whether computer vision is tracking the moment an item is out of stock or a scientist is scheduling equipment via their electronic lab notebook (ELN), connecting those data points into a central hub, like an advanced inventory management solution, ensures AI has the information it needs to create efficiencies in the lab.
It’s important to consider that proprietary applications and automation tools can prohibit connectivity. Work with a digital lab solutions provider that builds interoperability into their solutions, so labs can be scalable and flexible now and into the future.
- Leverage applications, connectivity, and automation to launch AI. When AI is integrated into an inventory management system, it combines the historical data with the real-time consumption changes collected by automation tools. The AI system can then analyze the data to uncover trends and insights for better operational decision making and more efficient operations.
Over time, machine learning within the AI will allow the system to become smarter, further fueling efficiency. AI systems themselves will also improve as the technology advances. For example, in the near future, a scientist could vocally communicate with generative AI as if it is a virtual assistant.
From smart shelves to inventory management solutions, an AI system enables real-time tracking and monitoring, automates routine manual tasks, and drives efficiencies across the full workflow. With the support of AI, lab managers can empower their teams to meet their operational challenges with enhanced efficiency — and fuel the discovery that gets life-changing treatments to market faster.
