Page 54 - CMA Journal (Jul-Aug 2026)

F oc us S ec t ion

Predicting Demand Before it Becomes a The important consideration is determining where Data silos, inconsistent formats and poor-quality Responsible AI Requires Human However, greater automation also increases the •
Smart Procurement for a Digital Economy: Problem automation adds value and where human approval information can signi cantly a ect the performance of Judgement importance of governance. Future procurement

must remain mandatory.
systems will need to balance agility and e ciency w
AI systems.

Harnessing AI and Machine Learning One of the strongest applications of AI in procurement Turning Spend Data into Strategic This makes data governance a procurement priority. AI should strengthen professional judgement rather sustainability, transparency, fairness and resilien
than replace it. Procurement decisions often involve
is predictive analytics. By analyzing historical
purchasing data, operational information and market Insight Organizations need integrated systems, clear data factors that cannot be reduced to historical data alone, Conclusion: Smarter Procurement, Not

trends, ML models can help forecast future demand. Procurement teams often hold signi cant amounts of standards and appropriate controls over data quality including ethics, strategic priorities, supplier Just Automated Procurement
and access.
relationships and exceptional circumstances. AI can
Procurement is Moving from what happened to Improved demand forecasting can help procurement data on suppliers, prices, purchasing patterns and For public-sector organizations, the challenge is even handle repetitive tasks, identify patterns and provide AI and ML have the potential to fundamentally •
market conditions. AI-based spend analytics can bring

Transactions to Strategy understanding why it teams avoid stock-outs, manage inventory more this information together to identify savings greater because data access must be balanced with recommendations, while procurement professionals strengthen procurement by improving e ciency,
happened, what is likely e ectively and improve supply-chain resilience. privacy, security and stewardship responsibilities. retain responsibility for interpreting those forecasting demand, identifying risks, supporting
The digital economy is changing how organizations to happen next and Research has highlighted the potential of intelligent opportunities, supplier risks and sourcing trends recommendations and making consequential supplier decisions and turning procurement data int
operate, compete and create value. Procurement is no what action should be procurement systems and big-data analytics to (Ilamurugan, 2026). The Risk of Automating Bias decisions (Tushar & Jaiswal, 2026). strategic insight.
longer simply an administrative function focused on taken. improve forecasting accuracy, while deep-learning Predictive analytics can help procurement AI does not automatically eliminate human bias. If Human-in-the-loop models are particularly relevant But technology alone is not the answer. Poor-qualit
processing purchase orders and controlling costs. It is technologies can support more detailed analysis of professionals move from analyzing historical historical procurement data contains biased decisions because they allow AI to provide recommendations data, algorithmic bias, regulatory uncertainty, pri
increasingly a strategic function that in uences Where AI Can material demand patterns (Sun et al., 2025). expenditure to anticipating future risks and or supplier preferences, AI systems may reproduce while keeping decision-making responsibility with concerns and organizational resistance can limit th
e ciency, resilience, supplier relationships, Make the Biggest The practical bene t is a shift from reactive purchasing opportunities. AI-based scenario planning can also those patterns. Research has raised concerns that authorized professionals. This approach is especially bene ts of intelligent procurement (Ayobami et
sustainability and organizational performance. Di erence to proactive planning. allow organizations to test di erent sourcing strategies AI-driven supplier scouting may result in increasingly important where procurement decisions can a ect 2024).

before implementing them. This is where procurement
Arti cial Intelligence (AI) and Machine Learning (ML) AI has applications Muhammad Musab Detecting Fraud and Managing Risk can move beyond being a cost centre and become a similar supplier pro les, potentially reducing supplier supply networks and wider stakeholder interests The future of procurement should therefore not be
are accelerating this shift by enabling procurement across almost every Mirza source of strategic intelligence. diversity and making it harder for new suppliers to (Ribeiro et al., 2025). de ned by how much human involvement can be
teams to process large volumes of data, identify stage of procurement, PhD Scholar in Procurement is particularly vulnerable to fraud, The Business Case: What Organizations enter the market (Ribeiro et al., 2025). Preparing the Procurement Workforce removed. It should be de ned by how e ectively

patterns and make faster, more informed decisions. including supplier Corporate Finance irregularities and con icts of interest. AI and ML can This creates what researchers describe as “bias Technology alone will not deliver digital procurement technology and professional expertise can work
These technologies can support smarter and more selection, demand forecasting, fraud detection, strengthen controls by identifying unusual transaction Can Gain co-evolution”, where historical patterns can in uence transformation. Organizations also need people who together. •
resilient procurement processes, although their contract management, spend analysis and strategic patterns, scoring supplier risks and monitoring The most immediate bene t of AI in procurement is future decisions and reinforce existing biases. understand how to use, question and govern AI For organizations operating in the digital economy,
adoption remains at an early stage. A review of 46 sourcing. A review by Spreitzenbarth et al. (2024) procurement activities in real time (Ezeji, 2024). e ciency. By automating repetitive activities, Organizations therefore need mechanisms to monitor systems. Successful AI adoption requires appropriate goal should be procurement that is faster, more
studies on AI and ML in Purchasing and Supply identi ed eleven major AI use-case clusters in Research into high-value public procurement has accelerating decision-making and reducing manual AI systems, identify problematic feedback loops and organizational infrastructure, cultural change and new transparent, more resilient and more sustainable
Management found that research has paid relatively procurement, including contract analysis, supplier and identi ed indicators such as single-bid tenders as e ort, AI can improve the speed and e ciency of intervene when necessary. workforce capabilities (Balkan & Akyüz, 2025). Many Achieving that goal requires more than AI and ML. I
limited attention to operational implementation, cost spend management, risk management, demand potential signals of corruption risk. Machine-learning procurement processes (Lakhal et al., 2025). The issue is particularly important in public organizations are still developing the expertise requires reliable data, capable people, clear

considerations and supplier sustainability forecasting and negotiation support. techniques, including Random Forest models, have AI can also improve decision quality by analyzing large procurement, where fairness, equal opportunity and required to design, implement and manage AI systems governance and responsible use of technology.
(Spreitzenbarth et al., 2024). also demonstrated potential in identifying patterns datasets and identifying patterns that may be di cult accountability are as important as e ciency. AI-driven e ectively.
Smarter Supplier Selection associated with procurement risks. for individuals to process manually. Decision-support When these elements come together, AI can help •
The opportunity, therefore, is not simply to automate systems can provide procurement professionals with decisions must therefore be subject to appropriate This is particularly important for the public sector, procurement evolve from a transaction-focused
procurement. It is to use technology to turn Choosing the right supplier directly a ects cost, For public-sector procurement, this capability can be timely information, predictive insights and scenario human oversight. where capacity constraints, complex procurement function into a strategic driver of value, risk
procurement data into practical intelligence and quality, delivery and long-term operational particularly valuable. Instead of relying solely on assessments (Paul et al., 2024). Governance Cannot Be An regulations and the need to balance nancial management, sustainability and organizational
strengthen its contribution to organizational value. performance. Traditional supplier assessment often periodic reviews, organizations can use AI to identify Afterthought e ciency with social and environmental objectives performance. •
depends heavily on human judgement and may potential anomalies earlier and direct human attention The bene ts extend beyond e ciency. AI can improve create additional challenges (Andersson et al., 2025).
What Procurement 4.0 Means in struggle to respond quickly to changing market towards transactions that require further investigation. visibility and traceability across procurement The rapid development of AI has moved faster than Investment in training and professional development References
Practice conditions. Automating Routine Procurement processes and support activities such as contract many regulatory frameworks. Organizations therefore should therefore be treated as an essential component • Andersson, P., Arbin, K., & Rosenqvist, C. (202
discovery, supplier matching, anomaly detection and
The concept of Procurement 4.0 brings Industry 4.0 AI can analyze both structured and unstructured Automation is another major opportunity. AI-powered tender evaluation (Patrucco et al., 2026; Hyppänen & face the challenge of adopting new technologies while of AI adoption rather than an optional activity. arti cial intelligence (AI) in governmental publi

Procurement,
120–139.
https://doi.org/10.1
25(1),
Public
technologies into procurement. AI, ML, the Internet of information to support supplier assessment and procurement systems can reduce repetitive Vilko, 2025). ensuring that procurement remains compliant, The Next Stage: From AI-Assisted to jopp-05-2024-0057
Things (IoT), blockchain and cloud computing can identify patterns that may not be immediately visible administrative work, accelerate processes and improve Ultimately, the strategic opportunity is to move transparent and accountable (Obinna & Kess-Momoh, AI-Enabled Procurement • Ayobami, A. T., Mike-Olisa, U., Ogeawuchi, J. C
support real-time monitoring, predictive analysis and to procurement teams. Machine-learning approaches, workforce productivity. The emergence of Agentic AI procurement from managing transactions to creating 2024). O. A. (2024). Digital Procurement 4.0: Redesignin
increasingly automated decision-making across the including Decision Trees, Random Forest and Gradient could take this further by enabling systems to perform value. AI can provide insights into supplier markets, Guidance such as IEEE P3119 provides complementary Procurement technology is continuing to evolve. Contracting Systems with AI-Driven Ethics, Compli
Optimization. International Journal of Scienti c

Source-to-Pay (S2P) process. risk exposure, sustainability and innovation process guidance for the procurement of AI and Agentic AI could enable systems to perform Science Engineering and Information Technology,
Boosting, are increasingly being explored for supplier tasks and make decisions within prede ned automated decision systems (Waters & Miller, 2025). increasingly autonomous tasks, while large language
This represents a signi cant change from traditional assessment. parameters. Such systems are being explored for opportunities, supporting broader organizational However, organizations need governance models may support conversational negotiations and https://doi.org/10.32628/cseit24102138
objectives.
procurement, where manual processes, fragmented activities including negotiation, compliance and arrangements that cover the entire AI lifecycle from other procurement interactions (Waditwar, 2025). • Balkan, D., & Akyüz, G. A. (2025). Arti cial in
learning (ML) in procurement and purchasing decis
data and individual judgement often dominated AI can also support the assessment of supplier nancial supplier management. The Data Problem: AI is Only as Good as selecting and deploying AI tools to monitoring their Digital twins could allow organizations to model taxonomic literature review and research opportun
decision-making. Digital systems can bring strength, risk pro les and other relevant information AI can also work alongside blockchain-based smart its Information performance. supply-chain scenarios before making procurement Intelligence Review, 58(11). https://doi.org/10.1
information together, identify patterns and provide through the analysis of large datasets. However, these contracts to trigger speci c contractual actions when decisions, while AI combined with blockchain, IoT and • Ezeji, C. L. (2024). Arti cial Intelligence for
insights across departments (Paul et al., 2024). tools should support not replace professional prede ned conditions are met, reducing The biggest obstacle to intelligent procurement may For procurement leaders, the question should not cloud technologies could create more connected and procurement fraud. International Journal of Busin
judgement, particularly where supplier selection has not be the technology itself, but the quality of the data simply be “Can we automate this?” but also “Should we Strategy (2687-2293), 6(1), 63–73. https://doi.or
AI and ML are particularly valuable because they can signi cant nancial or strategic consequences. administrative intervention and improving transaction behind it. AI and ML systems require reliable, complete automate this, under what controls, and who remains responsive procurement ecosystems (Özcan & Taş, • Hyppänen, T., & Vilko, J. (2025). AI in Public
move procurement beyond simply understanding e ciency. and accessible data. accountable?” 2026). Adoption in the Competitive Tendering Process. 1
10.1109/ice/itmc65658.2025.11106613
52 ICMA’s Chartered Management Accountant, Jul-Aug 2026