There is no doubt now that most of the planet is engulfed in a major inflation crisis. Prices on almost everything have gone up. While extreme cases, such as Venezuela, have seen inflation go up to over 1000% in recent periods, many developed countries are experiencing inflation at levels between 5-10%.
There are many causes to this, including oil price increases and the war in Ukraine, which are difficult to resolve in the near-term. On the one hand, spatial tools and GIS could offer a way to mitigate some of the impact of supply chains that are causing some of the high inflation prices, but longer-term we may need to think of new ways of configuring our supply chains.
Mapping Supply Chain Bottlenecks
At a basic level, GIS and spatial visualization has been used to map shipping and large container fleets around the world. Bottlenecks in supply chains, such as recently in Shanghai which is home to a lot of manufacturers important to global supply chains, caused by Covid-19 lockdowns are evident and can inform us that impending higher prices for various good are likely. The tool MarineTraffic can be used to visualize data using GPS receivers on most large container vessels.
While this is a simply way to keep track of where emerging bottlenecks are, there is a need to anticipate such bottlenecks well before they develop and find solutions to ease supply chain flows before it is too late.
One solution has been to use artificial intelligence as a way to anticipate supply chain disruptions and help find solutions. Tools that utilize Intelligent Spatial Decision Support Systems (SDSS) attempt to combine deep learning or general machine learning methods with GIS in order to trigger not only warnings that supply chain threats are emerging but to also seek alternative answers using a database of alternative manufacturers or supply networks that might be available. This could help ease immediate demand needs, even before full disruptions are evident.
Addressing Supply Chain Issues With Geospatial Methods
Researchers also see five main issues that supply chains need to address that spatial methods and machine learning could help resolve. These are supply chain networks, supplier selection, inventory planning, demand planning, and green supply chain management.
In the case of green supply chain management, this is mostly focusing on making choices that have less negative environmental impact. In procurement and supplier selection, there has been greater progress with recent software developed that industries are increasingly adopting. Artificial intelligence optimization in software, including examples such as ELI by Throughput Inc., Luminate by Blue Yonder, and llama.ai by Llamasoft, demonstrate that companies are beginning to utilize these tools to improve how they source materials based on emerging threats to supply chains.
Key to decision-making includes deployment of fuzzy sets theory and multi-criteria decision-making which factor social, economic, and environmental factors that could disrupt or affect supply networks, demand, and inventory.
Other works have been developing multi-item and multi-objective linear programming models that incorporate uncertainty in decision support. Methods have included using genetic algorithms that optimise based on different factors and using principals that apply to evolution, such as selection, to find optimal supply chain solutions.
Over the last seven years, perhaps the greatest expansion of research has been in using deep learning methods, such as in the Adaptive Neuro-Fuzzy Inference System, to solve multi-criteria decision-making problems and aid in decisions related to supplier evaluation and selection.
Easing Supply Chain Constraints
While current developments and tools are intended to help ease supply chain constraints, which would then help decrease inflation, clearly larger factors often prevent even the most intelligent supply chain systems from addressing all concerns.
Researchers have also been looking at wider strategic decisions, such as the effects of globalization, on affecting supply chains.
For decades, countries have been able to greatly benefit, particularly in the West, from fragmentation of supply chains and by geographical and sectoral concentration. This has led to more country specialization in global networks and enriched countries, but the cost has been, among other costs, is vulnerability to supply chains as they become dependent on different countries, particularly East Asia.
In the short-term, tools that help optimize supply chains networks as they exist might be the best solution, but larger strategic decision on potentially to either increase source countries for key products or create more local manufacturing in countries vulnerable to supply chain shocks might help prevent scenarios we are seeing today.
Other methods include using close loop supply chains that attempt to mitigate waste and create products from that waste, which has both environmental benefits and enables potential mitigation of supply chain stress.
Methods also try to optimize on geographic spread of needed products, such as in developing clean energy supply chains, where locating and building plants and manufacturing could mean less distance is needed between key products in supply chains.
Unfortunately, there is no easy way out of our current supply chain problems. New methods and tools have been created over the last decade that can ease some pressure by finding solutions using existing supply chain networks. However, our societies have not been developing robust supply chains but rather fairly fragmented ones that are prone to disruption. This means even the best methods cannot address current problems.
To resolve long-term issues, we may need to consider how to reconfigure supply chains so that future shocks are not as severe in affecting prices.
 For more on global inflation, see the Inflation Tracker by the Financial Times: https://www.ft.com/content/088d3368-bb8b-4ff3-9df7-a7680d4d81b2.
 For more on SDSS, see: R. Yusianto, Marimin, Suprihatin, and H. Hardjomidjojo, ‘Intelligent Spatial Decision Support System Concept in the Potato Agro-Industry Supply Chain’, in 2020 International Conference on Computer Science and Its Application in Agriculture (ICOSICA), Bogor, Indonesia, Sep. 2020, pp. 1–7. doi: 10.1109/ICOSICA49951.2020.9243233.
 For a review of the state of the art in supply chain analyses and systems, see: R. Sharma, A. Shishodia, A. Gunasekaran, H. Min, and Z. H. Munim, ‘The role of artificial intelligence in supply chain management: mapping the territory’, International Journal of Production Research, pp. 1–24, Feb. 2022, doi: 10.1080/00207543.2022.2029611.
 For more on how global supply chains have evolved in the last 15-20 years, see: S. Jiménez, E. Dietzenbacher, R. Duarte, and J. Sánchez-Chóliz, ‘The geographical and sectoral concentration of global supply chains’, Spatial Economic Analysis, pp. 1–25, Jan. 2022, doi: 10.1080/17421772.2021.2012584.
 For more on using spatial methods to optimize on supply chain locations for critical components, see: R. H. G. de Jesus, M. V. Barros, R. Salvador, J. T. de Souza, C. M. Piekarski, and A. C. de Francisco, ‘Forming clusters based on strategic partnerships and circular economy for biogas production: A GIS analysis for optimal location’, Biomass and Bioenergy, vol. 150, p. 106097, Jul. 2021, doi: 10.1016/j.biombioe.2021.106097.