Dr Marko Lange , SAP ’ s global head of chemicals , looks at the impact of artificial intelligence on the speciality chemicals supply chain
Surprise ! AI can help you plan for it
Dr Marko Lange , SAP ’ s global head of chemicals , looks at the impact of artificial intelligence on the speciality chemicals supply chain
Artificial intelligence ( AI ) cannot predict , months in advance , if , when , and where a hurricane may strike — and chaos theory reminds us of the impossibility of doing so . But it can help speciality chemical companies prepare their supply chains for greater or lesser surprises of all sorts , be they unpleasant or happy , meteorological , geopolitical , regulatory , economic or competitive .
Moreover , two of the hottest areas of AI interest in speciality chemicals , AI-powered product development and predictive quality , themselves have direct supply chain implications , as do demand planning , production planning , production control , quality control , natural-language chat capabilities and other AI-augmented systems and processes .
While diverse , these systems all share the ability to digest enormous , diverse quantities of structured and unstructured operational and business data as well as external data ( such as , for example , hurricane forecasts ), to draw connections and distil actionable conclusions for human intelligence to vet .
As AI speeds up decision-making , the supply chain must adapt apace . When planners change input assumptions based on evolving truths or speculative hunches , AI helps to understand the impact and find courses of action . Transportation management is a straightforward example of AI enhancing supply chain efficiency and resilience .
Conversational planning in the face of potential disaster
Take a hypothetical Atlantic hurricane , of which 11 are expected this year . Enhancing transportation management systems with AI lets supply chain planners run scenarios based on possible effects of different storm paths , because the impacts of a US Gulf Coast landfall will differ from one that strikes South Florida .
The system takes into account variables such as product type , transport capacity and costs , delivery reliability , lead time , projected revenue and sustainability . It then helps to find alternatives that propose varying transportation routes , different combinations of transportation modes ( ocean , rail , road , air ), and alternative production locations and then distils them down to a handful of recommended options .
That is already happening today . What is coming soon takes supply planning optimisation to an entirely new place . Generative AI ( GenAI ) - based conversational planning will let planners quickly find and vet sources of supply , check available stock , consider transport costs and constraints , evaluate supply chain risk and more .
The planner brings their knowledge of the situation at hand and likely courses of action . Based on the planner ’ s prompts , the system taps into diverse business and operational data to generate what-if scenarios the planner can then assess and refine ( again , conversationally ).
Importantly , GenAI can provide justification for the scenarios it deems to be most fruitful , providing the planner with a view into the infamous black box of AI decisionmaking . Or to put it another way , a key element of conversational planning is providing the human planner context and logical justification for the decisions AI is taking .
In our ‘ approaching hurricane ’ scenario , a planner will be able to ask
42 SPECIALITY CHEMICALS MAGAZINE ESTABLISHED 1981