Process-based roadmap for AI in procurement
| Procurement stage | AI applications | Enabling requirements | Key risks/constraints |
|---|---|---|---|
| S1. Demand and market analysis | Spend analysis, forecasting, market intelligence | Availability of structured historical procurement data; data integration across sources | Data quality issues; difficulty extracting relevant insights from large datasets |
| S2. Supplier selection and evaluation | Supplier scoring, risk assessment, decision support | Access to supplier data (internal/external); standardised evaluation criteria | Incomplete or inconsistent supplier data; limited transparency of AI outputs |
| S3. Contracting and negotiation | Contract analysis, support for negotiation processes | Digitised contractual data; formalised negotiation parameters | Legal and interpretative risks; limits of automation in complex negotiations |
| S4. Supplier management | Performance monitoring, anomaly detection, risk monitoring | Continuous data flows; KPI systems; data sharing with suppliers | Data silos; limited supplier integration; organisational resistance |
| S5. Strategy optimisation | Decision support, process optimisation, strategic analytics | Cross-functional data integration; analytical capabilities; organisational alignment | Lack of skills; unclear performance metrics; strategic misalignment |
| O1. Requirements definition | Demand prediction, automated planning | Integration with ERP/planning systems; real-time data inputs | Data inconsistency; system integration challenges |
| O2. Ordering | Order automation, RPA in purchasing processes | Standardised and digitised workflows | Automation errors; over-standardisation of processes |
| O3. Delivery monitoring | Tracking, delay prediction, logistics analytics | Access to real-time logistics data | Limited visibility; data latency |
| O4. Goods receipt and quality | AI-supported quality control (limited evidence) | Digital quality data and inspection systems | Very limited research coverage; underdeveloped applications |
| O5. Invoice and payment | Invoice automation, fraud detection (very limited evidence) | Structured financial and transactional data | Strong research gap; implementation uncertainty |
| Procurement stage | AI applications | Enabling requirements | Key risks/constraints |
|---|---|---|---|
| S1. Demand and market analysis | Spend analysis, forecasting, market intelligence | Availability of structured historical procurement data; data integration across sources | Data quality issues; difficulty extracting relevant insights from large datasets |
| S2. Supplier selection and evaluation | Supplier scoring, risk assessment, decision support | Access to supplier data (internal/external); standardised evaluation criteria | Incomplete or inconsistent supplier data; limited transparency of AI outputs |
| S3. Contracting and negotiation | Contract analysis, support for negotiation processes | Digitised contractual data; formalised negotiation parameters | Legal and interpretative risks; limits of automation in complex negotiations |
| S4. Supplier management | Performance monitoring, anomaly detection, risk monitoring | Continuous data flows; KPI systems; data sharing with suppliers | Data silos; limited supplier integration; organisational resistance |
| S5. Strategy optimisation | Decision support, process optimisation, strategic analytics | Cross-functional data integration; analytical capabilities; organisational alignment | Lack of skills; unclear performance metrics; strategic misalignment |
| O1. Requirements definition | Demand prediction, automated planning | Integration with ERP/planning systems; real-time data inputs | Data inconsistency; system integration challenges |
| O2. Ordering | Order automation, RPA in purchasing processes | Standardised and digitised workflows | Automation errors; over-standardisation of processes |
| O3. Delivery monitoring | Tracking, delay prediction, logistics analytics | Access to real-time logistics data | Limited visibility; data latency |
| O4. Goods receipt and quality | AI-supported quality control (limited evidence) | Digital quality data and inspection systems | Very limited research coverage; underdeveloped applications |
| O5. Invoice and payment | Invoice automation, fraud detection (very limited evidence) | Structured financial and transactional data | Strong research gap; implementation uncertainty |
Sharing content requires targeting cookies to be enabled. Please update your cookie preferences to use this feature.