DAL.IA: Using Artificial Intelligence to Manage the Supply Chain
Mar 30, 2022 11:00:00 AM
A conversation with Mrs. Anne Nguyen, Product Director, Analytics and Innovation at Logibec, who is spearheading the development of this product.
1 - We are speaking today to discuss the DAL.IA project that you are leading, in collaboration with the Centre hospitalier de l'Université de Montréal (CHUM), to support procurement and performance at hospitals in Québec. Could you tell me about this project?
"DAL.IA is a project for teams that are not usually top of mind when hospitals decide to integrate artificial intelligence into their operations. Historically, in response to significant increases in operational costs, the healthcare network has relied on cost-saving solutions based on obtaining materials at the lowest price, consolidating purchases, and managing minimum/maximum stock levels, which consists in replenishing stocks to the maximum level when they reach a minimum threshold. It is clear that this model, while producing what appears to be short-term savings, does not eliminate the possibility of stock-outs - which in turn lead to a number of potentially cumulative consequences. First and foremost, by undermining the trust that healthcare providers and patients have in the facility regarding supply availability, but also, by creating a need for emergency workarounds, which are often quite costly.
Logibec continues to expand its procurement expertise by further investing in the development of tools that are better adapted to technological changes, geographic constraints, and the growing demand for care delivery. Our Financial and Material Management solution (FMMS) meets the operational and financial efficiency needs of all institutions in the Québec healthcare network. It is appreciated for its high level of interoperability with peripheral tools and its flexibility in adapting to each facility’s unique administrative structure. In recent years and months, Logibec's team of experts has focused its efforts on creating a predictive module powered by artificial intelligence. This module, DAL.IA, puts artificial intelligence at the service of the supply chain. DAL.IA takes us from the "action-reaction" era to the "prediction-action" era. The tool will allow administrators to avoid emergency situations by optimizing purchasing."
2 - What prompted this project?
"The exponential increase in demand for both acute and long-term care is resulting in ever-increasing costs. To date, the system has focused on short-term cost savings, but it is now becoming apparent that this cost-saving model is not sustainable, as it does not consider the actual level of use of each product. Seasonality, context-driven quality, and waste levels must be diligently evaluated and calculated to effect real change. Currently, there are no technological tools to predict the risk of inventory shortages. However, inventory shortages can have costly consequences for a hospital, particularly when they result in cancelled surgeries. This is unfortunately a reality. Eliminating unnecessary out-of-pocket expenses is a major concern, but patients remain our priority. All hospitals cancel a significant percentage of surgeries due to the inability to keep operating rooms adequately stocked. We are all sensitive to this. DAL.IA's predictive analytics will ensure inventory availability and have a positive effect on decreasing cancellation rates at facilities.
To develop the pilot project, Logibec brought in two major partners. The first is the Centre hospitalier de l'Université de Montréal (CHUM). Known for its desire to contribute to the evolution of the healthcare network, particularly through artificial intelligence, and given the volume of annual transactions and products in daily circulation in the facility, the CHUM offers ideal conditions for developing a project like DAL.IA. This facility has also been using our FMMS solution for several years. In fact, the institution's procurement process is highly digitized, which will allow the team not only to collect a large amount of information, but also to work with real, quality data.
The second partner is Scale AI, a central pillar of the artificial intelligence ecosystem in Canada. The company's mission is to foster technological excellence by supporting innovative development and breakthrough technologies. Scale AI's support in this project is a testament to the relevance and merit of the product."
3 - What are some of the challenges or obstacles you are facing?
"As part of this pilot project, we chose to develop two modules: delivery status prediction and usage prediction. Delivery status prediction identifies the risk that an order line will be incomplete or late when the next order is issued. The usage prediction component, on the other hand, predicts variations in usage, anomalies, and quantities used over time. There is no such thing as zero risk, but we opted to implement a project with a low level of risk. In an artificial intelligence project, the main issues are the quantity and quality of the data, so we chose partners with extensive expertise in data science. So far, we have not identified any risks with respect to the pool of data available to us for this pilot project. Of course, we need to separate the "COVID" data from the so-called historical data to ensure that prediction models will match real-life situations."
4- Acknowledging that the project is still in development, can you share some of your predictions? What results are you hoping to see once DAL.IA has been implemented?
"We have high aspirations for the DAL.IA project. Supply chain-related costs are one of the largest expenses in the healthcare system and can account for as much as one-third of total operating expenses for some hospitals. In total, this represents approximately $3 billion for the province. Let's think back to the cancelled surgeries we mentioned earlier. Ariane Lacoursière, a journalist at La Presse, stated in April 2016 that, at the CHUM, approximately 6% of surgeries had been cancelled. Of this percentage, 19% were cancelled due to inventory shortages. In addition, recent events such as natural disasters and infectious disease outbreaks have highlighted the challenges posed by disruptions in the continuity of supply.
The first objective, targeted with the implementation of DAL.IA, is to ensure continuity of supply, regardless of context, and to reduce additional costs. As a result, facilities will see greater fluidity across all their activities. From the operating room to the kitchen, through every technical platform, all departments and care units will benefit. Patients will also benefit since they are at the very end of the supply chain.
Savings are not only found in streamlined and more cost-effective purchasing: they are also generated by shorter lengths of stay, in every case where a patient's stay might have been extended due to a delay or disruption in the supply chain. A bed that does not become available at the originally scheduled time delays all other admissions. This can jeopardize the safety of some patients.
Beyond these quantifiable considerations, Logibec also cares about the satisfaction of patients, clinical staff, and supply chain stakeholders. Automating processes to eliminate manual and repetitive tasks is a form of artificial intelligence. Automation is being used more and more because it frees healthcare professionals from administrative overhead so they can spend more time with patients. We are convinced that DAL.IA has the potential to become an essential tool for hospital management throughout Québec."
5 - What are the next steps before going to market?
"We will develop a proof of concept with our partners, which will allow us to address the needs of the entire healthcare network, and also approach the Canadian, American, and European markets. The teams are complete. We have obtained all the approvals we need. The next steps will involve infrastructure, data pools and data security. These are very technical steps."
6 - What are your recommendations to manage change within a facility or, more precisely, within a purchasing department?
"First, we need to ensure that end-users have a basic understanding of the concept of "prediction". This is essential to achieve high adoption of the product and, above all, for the process to unfold smoothly. It is therefore important to involve stakeholders along the supply chain as early as possible to allow them to appropriate the tool gradually. The Information Technology (IT) department must also be included in the development of the solution, as its employees must quickly master the tool once it is ready to be deployed. The success of such an initiative depends on the involvement of all stakeholders. A training plan must also be developed to boost the adoption rate."
7 - In your opinion, why is innovation essential for the healthcare network in Québec?
"Logibec sees infinite potential for integrating artificial intelligence into the activities of the healthcare network. All aspects of hospital management present exciting opportunities for innovation supported by artificial intelligence. Just think of surgical backlog management, a hot topic that lends itself perfectly to the development of artificial intelligence tools. And what about virtual care platforms that adapt to both the needs and preferences of patients to create customized monitoring, recovery, rehabilitation, or prevention programs? These will eventually be part of the post-pandemic solution! There are so many innovative ideas to tap into, especially in the current context, where the healthcare network is struggling, but ripe for deep transformation. DAL.IA will be our first breakthrough in artificial intelligence, but we won't stop there! DAL.IA will bring us legitimacy and recognition in the small world of artificial intelligence, allowing us to go even further. What motivates our troops in the development of technological tools is improving well-being, safety, comfort, and accessibility to care for patients and staff alike."
The DAL.IA module will be offered with version 8.0 of our Financial and Material Management solution (FMMS). Contact a member of the Logibec team to set up your progressive upgrade plan. The FMMS product upgrade is performed while maintaining the continuity of your assets. Get started and benefit from modern, effective, flexible and user-friendly tools.
 Data obtained by the journalist through the Access to Information Act.