Digital supply chain surveillance using artificial intelligence: definitions, opportunities and risks

Figures & data

Table 1. Supply chain surveillance challenges.

ID	Surveillance challenge – example	Category	Stakeholders	Relevant references that mention challenge category	Approaches proposed to address DSCS Challenges	Data sources used	AI methods used
A	If I order part a, from supplier s, it is likely to arrive 3 days late	Resilience	Buyer	Harland, Brenchley, and Walker (2003), Blackhurst, Wu, and O’Grady (2004); Manuj and Mentzer (2008), Tang and Tomlin (2008), Tang and Nurmaya Musa (2011), Tummala and Schoenherr (2011), Samvedi et al. (2013)	He et al. (2014), Brintrup et al. (2020), Zheng, Kong, and Brintrup (2023)	ERP data	Classification (Naïve Bayes, Support Vector Machine, Gradient Boosting, Decision Trees) Regression (Neural network)
B	Supplier is disrupted with likelihood	Resilience	Buyer, Insurer	Harland, Brenchley, and Walker (2003), Blackhurst, Wu, and O’Grady (2004); Manuj and Mentzer (2008), Tang and Tomlin (2008), Tang and Nurmaya Musa (2011), Tummala and Schoenherr (2011), Samvedi, Jain, and Chan (2013)	He et al. (2014), O’Leary (2015)	Social media	Natural Language Processing
C	Supplier has quality issues for product	Product quality	Buyer	Blackhurst, Scheibe, and Johnson (2008)	Psarommatis et al. (2020)	ERP data, maintenance logs	Classification (Neural Networks)
D	Buyer b is likely to be connected to suppliers in a hazardous zone	Resilience	Buyer, Insurer	Jüttner, Peck, and Christopher (2003), Lin and Zhou (2011)	Brintrup et al. (2018), Xie et al. (2019); Aziz et al. (2021), Kosasih and Brintrup (2021a), Wichmann et al. (2018), Kosasih and Brintrup (2021b)	World Wide Web, industrial databases, annual reports	Natural Language Processing, Classification (Neural Networks) Knowledge Graph Completion, Neuro Symbolic AI
E	Supplier s might be supplier to buyer b	Visibility, Resilience, Competition	Buyer, Insurer	Jüttner, Peck, and Christopher (2003), Lin and Zhou (2011), Harland, Brenchley, and Walker (2003)	Brintrup et al. (2018), Xie et al. (2019); Aziz et al. (2021), Kosasih and Brintrup (2021a); Wichmann et al. (2018), Kosasih and Brintrup (2021b)	World Wide Web, industrial databases, annual reports	Natural Language Processing, Classification (Neural Networks) Knowledge Graph Completion, Neuro Symbolic AI
F	Supplier s might be connected to supplier k	Visibility, Resilience	Buyer, Insurer	Jüttner, Peck, and Christopher (2003), Lin and Zhou (2011)	Brintrup et al. (2018), Xie et al. (2019); Aziz et al. (2021), Kosasih and Brintrup (2021a); Wichmann et al. (2018), Kosasih and Brintrup (2021b)	World Wide Web, industrial databases, annual reports	Natural Language Processing, Classification (Neural Networks) Knowledge Graph Completion, Neuro Symbolic AI
G	My product p is likely to contain nuts	Product quality	Buyer, Insurer	Harland, Brenchley, and Walker (2003)	Ahn et al. (2011)	Food retail datasets	Classification (Neural Networks), Clustering
H	This is a counterfeit product	Product quality	Buyer, Regulatory	Harland, Brenchley, and Walker (2003)	Ahmadi, Javidi, and Shahbazmohamadi (2018)	N/A	Classification (Neural Networks)
I	Supplier is unsustainable	Sustainability	Buyer, Insurer	Samvedi, Jain, and Chan (2013), Harland, Brenchley, and Walker (2003)	N/A	Carbon emission reporting	Classification (Neural Networks)
J	We cannot lend to Suppliers because it sells to disreputable buyer b	Financial	Financer	Harland, Brenchley, and Walker (2003)	Martinez et al. (2019)	Financial records, World Wide Web, industrial databases, annual reports	Natural Language Processing
K	Supplier s might have financial problems	Financial	Buyer, Financer	Harland, Brenchley, and Walker (2003)	Martinez et al. (2019),	Financial records	Natural Language Processing
L	Buyer b might have financial problems	Financial	Supplier, Financer	Harland, Brenchley, and Walker (2003)	n/a	Financial records	Natural Language Processing
M	Supplier has excess capacity	Negotiation	Buyer	n/a	n/a	Shipment records, ERP data
N	Supplier is likely to offer high price for this product	Negotiation	Buyer	n/a	Jiao, You, and Kumar (2006), Boateng et al. 2017	Spenditure data	Classification
O	Supplier is innovative	Innovation	Buyer	Aristodemou et al. (2018), Trautrims et al. (2017)	Aristodemou et al. (2018), Trautrims et al. (2017)	Patents	Neural networks
P	Buyer is not likely to accept this bid for this product	Negotiation	Supplier	Yang and Sun (2019)	Yang and Sun (2019)	Sales records	Multi-agent learning

Table 2. Current commercial solutions offered for DSCS.

Company	Category	Description
Sourcemap https://sourcemap.com	Visibility, Resilience, Sustainability	Tool to ask suppliers to share sub-tier information, which then allows a snowballing process. DSCS concentrates on the application of AI to automatically query public compliance databases and proprietary risk databases
Resilinc https://www.resilinc.com	Visibility, Resilience	EventWatch solution that monitors news for a given set of suppliers, and a multi-tier mapping solution which collects data from the web, bills of lading, and press releases
Everstream Analytics https://www.everstream.ai/	Visibility, Resilience, Sustainability	News monitoring with Natural language processing, connects to ERP data, and proprietary data sources
ImportGenius https://www.importgenius.com	Negotiation	Allows tracking of competitor’s shipments and supplier to identify prospect buyers. Allows companies to monitor public companies’ imports and exports and identify market trends and predict on the success or failure of new product launches. Uses ocean freight records
Altana AI https://www.altana.ai	Visibility, Resilience	Attempts to federate different data streams from shipments and bills of lading with HS codes, and extracts company relationships with Natural Language Processing. Uses predictive algorithms for entity resolution (which company does a given shipment record refer to) and predicting what is inside a shipment container and its volume and associated value. Resolves time lags with ‘nowcasting' to predict who supplies whom. Provides a shipment rating for customs compliance. Founders are former-Panjiva employees
Panjiva https://panjiva.com	Visibility, Resilience	Uses government data sources with freedom of information requests. Uses shipment data, public and private subscription company data, port of lading data. Acquired by S&P Global
S&P Global Market Intelligence https://www.spglobal.com/	Financial	Prediction of financial health, using proprietary data sources. Uses Panjiva’s technology to report n Supply Chain Intelligence
VersedAI https://www.versed.ai	Visibility, Resilience	Built on technology initially proposed by Wichmann et al. (2018). Uses publicly available web data to train classifiers and automatically maps supply chains
HICX https://www.hicx.com	Resilience, Sustainability	Solution to consolidate Enterprise-wide supplier data with ERP based performance data to make predictions on supplier experience and performance
MakerSite https://makersite.io	Visibility, Resilience, Sustainability	After obtaining product composition and first tier supplier datasets, connects to proprietary databases to obtain sub-tier information
FRDM https://www.frdm.co	Resilience	Uses purchasing data from a company’s ERP, to automatically map the first tier supply chain. Juxtaposes media alerts on the extracted map for ongoing risk alerts using Natural language processing tools
OpenSC https://opensc.org	Sustainability	Focussed in food production. Verifies ethical and sustainable practices on the supply chain such as fair payments to farmers, fishing sustainably. Uses block chain and RFID scans to collect information so not entirely independent of supplier consent. The use of DSCS technology concerns the verification stage with GPS tracking. Here if a vessel was fishing in a protected area by looking at its GPS locations, coordinated with protected zones, the vessel’s speed, sea depth

Table 3. Regression models.

Model type	Dependent variable	Independent Variables
BLR – Model 1	Use of internal datasets	Sector, Size, Complexity*, International, Visibility
BLR – Model 2	Use of social media	Sector*, Size, Complexity, International, Visibility
BLR – Model 3	Use of supplier websites	Sector, Size^., Complexity*, International, Visibility
BLR – Model 4	Use of search engine	Sector, Size, Complexity, International, Visibility**
BLR – Model 5	Use of external audit companies	Sector, Size, Complexity^., International**, Visibility^.
BLR – Model 6	Use of business intelligence software	Sector^., Size^., Complexity^., International, Visibility^.
BLR – Model 7	Use of artificial intelligence	Sector*, Size, Complexity, International, Visibility
OLR – Model 1	Level of automation	Sector*, Size^., Complexity^., International, Visibility
OLR – Model 2	Proportion of intelligence gathering on suppliers	Sector*, Size, Complexity, International, Visibility^.
OLR – Model 3	Regularity of intelligence gathering	Sector*, Size, Complexity^., International, Visibility
OLR – Model 4, 5, 6	Delays, Excess Capacity, Quality, Price	Sector*, Size, Complexity^., International, Visibility
BLR – Model 8	Explainability (overall)	Sector*, Size*, Complexity*^., International, Visibility*
BLR – Model 9	Explainability (supplier performance)	Sector*, Size, Complexity, International, Visibility*
BLR – Model 10	Explainability (demand forecasting)	Sector*, Size, Complexity, International, Visibility
BLR – Model 11	Explainability (delay prediction)	Sector*, Size, Complexity, International, Visibility*

. = P < 0.1, *= P < 0.05, **= P < 0.01, ***= P < 0.001

Figure 1. Supply chain properties of companies that were surveyed.

Figure 2. Illustration of digital supply chain surveillance survey responses.

Table 4. Example sentences extracted from Reuters text corpus.

The Boeing Co. selected UTC Aerospace Systems to work with it to develop advanced actuation technology for the U.S. Air Force.

Hexcel will extend its partnership with Airbus Helicopters by supplying composite materials for the new H160 helicopter.

In 2014, RUAG and Tata Group signed an agreement for the latter to become a key supplier of the program

Advanced composite manufacturer, Teledyne CML Composites is celebrating the award by GKN Aerospace of a package of composite nacelle parts

Boeing Co has entered into a $1-billion agreement with Russian titanium producer VSMPO-Avisma Corp., to source various titanium products from 2011 through 2015

Figure 3. (a) Knowledge Graph before link prediction and (b) Knowledge graph after link prediction.

Table 5. Degree centrality before and after link prediction.

Before		After
Company	Degree Centrality	Company	Degree Centrality
Boeing	0.221	Boeing	0.521
Airbus	0.147	Airbus	0.336
Toyota	0.063	Toyota	0.063
GM	0.023	Rolls-Royce	0.053
Rolls-Royce	0.023	Honeywell	0.041
Apple	0.023	Thales	0.037
BMW	0.023	Rockwell Collins	0.033
GKN Aerospace	0.019	Lockheed Martin	0.028
Embraer	0.016	GKN Aerospace	0.027
Rockwell Collins	0.014	GM	0.027

Table 6. Eigenvector centrality before and after link prediction.

Before		After
Company	Eigenvector Centrality	Company	Eigenvector Centrality
Boeing	0.594	Boeing	0.510
Airbus	0.366	Airbus	0.398
Rolls-Royce	0.095	Rolls-Royce	0.109
Rockwell Collins	0.085	Honeywell	0.090
UTC Aerospace Systems	0.079	Thales	0.080
BAE Systems	0.076	Rockwell Collins	0.080
Alcoa	0.074	Lockheed Martin	0.027
GKN	0.072	Mitsubishi Motors Corp	0.068
Spirit AeroSystems	0.072	Chrysler	0.067
FACC	0.071	Bombardier	0.064

Table 7. Betweenness centrality before and after link prediction.

Before		After
Company	Betweenness Centrality	Company	Betweenness Centrality
Boeing	0.337	Boeing	0.643
Airbus	0.226	Airbus	0.234
Toyota	0.111	Toyota	0.075
BMW	0.054	Apple	0.025
Kobe Steel	0.045	Kaiser	0.021
Alcoa	0.043	General Motors	0.019
Apple	0.038	GKN Aerospace	0.018
GM	0.037	Pratt & Whitney	0.016
General Motors	0.035	BMW	0.015
GKN	0.034	Foxconn	0.015

Table 8. Closeness centrality before and after link prediction.

Before		After
Company	Closeness Centrality	Company	Closeness Centrality
Boeing	0.340	Boeing	0.593
Airbus	0.322	Airbus	0.510
Kobe Steel	0.279	Rolls-Royce	0.404
GKN	0.278	Kobe Steel	0.402
Bridgestone	0.276	Tenneco	0.400
ASCO	0.275	Thales	0.399
Hexcel	0.274	Rockwell Collins	0.399
Alcoa	0.273	Yazaki	0.398
Eaton Corporation	0.267	GM	0.398

Harland, C., R. Brenchley, and H. Walker. 2003. “Risk in Supply Networks.” Journal of Purchasing and Supply Management 9 (2): 51–62. https://doi.org/10.1016/S1478-4092(03)00004-9

 Google Scholar

Blackhurst, J., T. Wu, and P. O’Grady. 2004. “Network-based Approach to Modelling Uncertainty in a Supply Chain.” International Journal of Production Research 42 (8): 1639–1658. https://doi.org/10.1080/0020754030360001646064

 Web of Science ®Google Scholar

Manuj, I., and J. T. Mentzer. 2008. “Global Supply Chain Risk Management.” Journal of Business Logistics 29 (1): 133–155. https://doi.org/10.1002/j.2158-1592.2008.tb00072.x

 Web of Science ®Google Scholar

Tang, C., and B. Tomlin. 2008. “The Power of Flexibility for Mitigating Supply Chain Risks.” International Journal of Production Economics 116 (1): 12–27.

 Web of Science ®Google Scholar

Tang, O., and S. Nurmaya Musa. 2011. “Identifying Risk Issues and Research Advancements in Supply Chain Risk Management.” International Journal of Production Economics 133 (1): 25–34. https://doi.org/10.1016/j.ijpe.2010.06.013

 Web of Science ®Google Scholar

Tummala, R., and T. Schoenherr. 2011. “Assessing and Managing Risks Using the Supply Chain Risk Management Process (SCRMP).” An International Journal 16 (6): 474–483.

 Web of Science ®Google Scholar

Samvedi, A., V. Jain, and F. T. Chan. 2013. “Quantifying Risks in a Supply Chain through Integration of Fuzzy AHP and Fuzzy TOPSIS.” International Journal of Production Research 51 (8): 2433–2442.

 Web of Science ®Google Scholar

He, M., H. Ji, Q. Wang, C. Ren, R. M. Lougee, H. Ji, Q. Wang, C. Ren, and R. Lougee. 2014. “Big Data Fueled Process Management of Supply Risks: Sensing, Prediction, Evaluation and Mitigation.” Proceedings of the 2014 Winter Simulation Conference : 1005–1013.

 Google Scholar

Brintrup, A., J. Pak, D. Ratiney, T. Pearce, P. Wichmann, P. Woodall, and D. Mcfarlane. 2020. “Supply Chain Data Analytics for Predicting Supplier Disruptions: A Case Study in Complex Asset Manufacturing.” International Journal of Production Research 58 (11): 3330–3341. https://doi.org/10.1080/00207543.2019.1685705

 Web of Science ®Google Scholar

Zheng, G., L. Kong, and A. Brintrup. 2023. “Federated Machine Learning for Privacy Preserving, Collective Supply Chain Risk Prediction.” International Journal of Production Research, 1–18. https://doi.org/10.1080/00207543.2022.2164628

 Web of Science ®Google Scholar

O’Leary, D. E. 2015. “Twitter Mining for Discovery, Prediction and Causality.” Intelligent Systems in Accounting, Finance and Management 22 (3): 227–247.

 Google Scholar

Blackhurst, J. V., K. P. Scheibe, and D. J. Johnson. 2008. Supplier risk assessment and monitoring for the automotive industry. International journal of physical distribution and logistics management 38 (2): 143–165.

 Google Scholar

Psarommatis, F., G. May, P.-A. Dreyfus, and D. Kiritsis. 2020. “Zero Defect Manufacturing: State-of-the-art Review, Shortcomings and Future Directions in Research.” International Journal of Production Research 58 (1): 1–17. https://doi.org/10.1080/00207543.2019.1605228

 Google Scholar

Jüttner, U., H. Peck, and M. Christopher. 2003. “Supply Chain Risk Management: Outlining an Agenda for Future Research.” International Journal of Logistics: research and applications 6 (4): 197–210.

 Google Scholar

Lin, Y., and L. Zhou. 2011. “The Impacts of Product Design Changes on Supply Chain Risk: A Case Study.” International Journal of Physical Distribution and Logistics Management 41 (2): 162–186.

 Web of Science ®Google Scholar

Brintrup, A., P. Wichmann, P. Woodall, D. Mcfarlane, E. Nicks, and W. Krechel. 2018. Predicting Hidden Links in Supply Networks. Complexity; Hindawi. https://doi.org/10.1155/2018/9104387

 Google Scholar

Xie, M., T. Wang, Q. Jiang, L. Pan, and S. Liu. 2019. “Higher-Order Network Structure Embedding in Supply Chain Partner Link Prediction.” In Computer Supported Cooperative Work and Social Computing, 3–17. Springer.

 Google Scholar

Aziz, A., E. Kosasih, R. Griffiths, and A. Brintrup. 2021. Data Considerations in Graph Representation Learning for Supply Chain Networks. International Conference on Machine Learning.

 Google Scholar

Kosasih, E., and A. Brintrup. 2021a. A Machine Learning Approach for Predicting Hidden Links in Supply Chain with Graph Neural Networks. https://10/325126

 Google Scholar

Wichmann, P., A. Brintrup, S. Baker, P. Woodall, and D. McFarlane. 2018. “Towards Automatically Generating Supply Chain Maps from Natural Language Text.” IFAC-PapersOnLine 51 (11): 1726–1731. https://doi.org/10.1016/j.ifacol.2018.08.207

 Google Scholar

Kosasih, E. E., and A. Brintrup. 2021b. Reinforcement Learning Provides a Flexible Approach for Realistic Supply Chain Safety Stock Optimisation. ArXiv:2107.00913 [Cs]. http://arxiv.org/abs/2107.00913

 Google Scholar

Ahn, Y. Y., S. E. Ahnert, J. P. Bagrow, and A.-L. Barabási. 2011. “Flavor Network and the Principles of Food Pairing.” Scientific Reports 1 (1): 196. https://doi.org/10.1038/srep00196

 PubMedGoogle Scholar

Ahmadi, B., B. Javidi, and S. Shahbazmohamadi. 2018. “Automated Detection of Counterfeit ICs Using Machine Learning.” Microelectronics Reliability 88–90: 371–377. https://doi.org/10.1016/j.microrel.2018.06.083

 Web of Science ®Google Scholar

Martínez, A., J. Nin, E. Tomás, and A. Rubio. 2019. “Graph Convolutional Networks on Customer/Supplier Graph Data to Improve Default Prediction.” In Complex Networks X, edited by S. P. Cornelius, C. Granell Martorell, J. Gómez-Gardeñes, and B. Gonçalves, 135–146. Springer International Publishing. https://doi.org/10.1007/978-3-030-14459-3_11

 Google Scholar

Jiao, J. R., X. You, and A. Kumar. 2006. “An Agent-Based Framework for Collaborative Negotiation in the Global Manufacturing Supply Chain Network.” Robotics and Computer-Integrated Manufacturing 22 (3): 239–255. https://doi.org/10.1016/j.rcim.2005.04.003

 Web of Science ®Google Scholar

Aristodemou, L., and F. Tietze. 2018. “The State-of-the-art on Intellectual Property Analytics: A Literature Review on Artificial Intelligence, Machine Learning and Deep Learning Methods for Analysing Intellectual Property (IP) Data.” World Patent Information 55: 37–51. https://doi.org/10.1016/j.wpi.2018.07.002

 Web of Science ®Google Scholar

Trautrims, A., B. L. Maccarthy, and C. Okade. 2017. “Building an Innovation-Based Supplier Portfolio: The Use of Patent Analysis in Strategic Supplier Selection in the Automotive Sector.” International Journal of Production Economics 194: 228–236. https://doi.org/10.1016/j.ijpe.2017.05.008

 Web of Science ®Google Scholar

Yang, C., and J. Sun. 2019. “Research on Negotiation of Manufacturing Enterprise Supply Chain Based on Multi-Agent.” Journal of Internet Technology 20 (2): 389–398.

 Web of Science ®Google Scholar

Data availability statement

Due to the nature of the research, supporting data is not available.