
Published
Author
Pharmaceutical supply in the Nordics is complex. But as new technology can change what can be predicted, automated and managed, we also need to dare to rethink the processes, ways of working and systems on which pharmaceutical supply is based.
Ensuring that the right medicine is in the right place when the patient needs it should be an obvious objective. In practice, the journey from production to patient is considerably more complicated. For example, a medicine may be produced and certified for release by a Qualified Person in one country, transported for storage with an external logistics partner in another, and then distributed from there to several Nordic markets. For each market, the medicine must be approved and meet the applicable regulatory and commercial requirements before it can be made available for sale.
On top of this, Good Distribution Practice (GDP) imposes stringent requirements for quality, traceability and documentation, as well as the varying regulatory requirements in the Nordic countries for different types of medicine. Many life sciences companies also work with several contract manufacturing organisations (CMOs), third-party logistics providers (3PLs), wholesalers and other external parties, while critical information may be scattered across different departments, held in different systems or stored in various manual solutions.
The result is complex arrangements that were originally created to quickly solve a specific, limited problem. When information and expertise become dependent on individuals, vulnerability increases. Instead, the organisation needs to build coherent, sustainable and value-creating supply chains that work over time.
AI and other emerging technologies create opportunities to move from reactive problem-solving to more proactive management. With AI and better-integrated data, changes in demand, stock levels and delivery risks can be identified earlier, giving the organisation a better chance to act before a shortage occurs. This could mean moving from asking, “Why did we run short of this product?” to asking, “Where are we at risk of a shortage, and what can we do about it before it happens?”
For life sciences, areas such as demand forecasting, inventory optimisation, vendor risks, supply chain visibility and predictive alerts could all present new opportunities through AI and better-integrated data. But the potential goes far beyond better forecasting. Life sciences often need to rethink the entire process. The really interesting question is not how AI can be added on top of today’s processes, but how those processes would be designed if you had the opportunity to build them from scratch using today’s technology.
Life sciences companies need to ask questions such as:
Roles in supply planning will change. Planners will need to spend less time collecting and compiling information, and more time assessing deviations, scenarios and consequences. AI can detect patterns, identify deviations and provide a better basis for decision-making. People contribute operational knowledge, experience and accountability for the decisions made. The true potential lies in using technology to enhance people’s capabilities, not to replace them.
In a regulated life sciences operation, greater efficiency must never mean reduced control. New technology therefore needs to be combined with clear areas of responsibility, high data quality, traceability, quality assurance and the right level of human oversight. Disruptions to production and logistics will still occur. The difference is that better data and AI can help people identify risks earlier, understand the consequences and act together with their partners before problems escalate.
It doesn’t start with which system the company should buy. When organisations see a need for better planning or control, it is easy to start discussing ERP systems, APS, WMS, Control Towers or AI solutions straight away.
But technology should not be the starting point. Instead, start with the business:
What does our supply chain look like today? Where are the risks and bottlenecks? Which activities actually create value? What information do we need but currently lack? What can be simplified or automated? And how do we want the process to work in the future? Only then does the question of what system support is needed become relevant.
An ERP system may be sufficient in some areas. In others, it may need to be supplemented with APS or other specialist solutions, for example. The future of pharmaceutical supply lies in the interplay between people and AI. The opportunity lies in combining life sciences and supply chain expertise with process development, data and modern technology to build simpler, safer, more proactive and sustainable supply chains.
In summary, then, the first step does not necessarily need to be buying a new system. Instead, start by creating a clear picture of today’s supply chain: the processes, ways of working, information flows, system support and biggest risks. By identifying where change can create the greatest value, you will also move one step closer to a supply chain that ensures the right medicines are in the right place at the right time, with full traceability and control.
1. Start with the process, not the technology
Always start with an analysis of the current state of processes, working methods and system support.
2. Identify where the risks lie
Map bottlenecks, manual tasks and critical dependencies on individual people.
3. Ensure access to the right data
Review data quality and structure, including packaging data and relevant ESG data.
4. Define how technology and people work together
Identify what technology can do and where human judgement, responsibility and control are needed.
5. Develop processes and system support together
Start with how you want the future supply chain to work, then define the system support required.
With expertise in life sciences, supply chain, process development and systems, HerbertNathan & Co helps you identify risks, development opportunities and the path towards a more proactive and sustainable supply chain.