E-ISSN:2583-2468

Research Article

Generative AI

Applied Science and Engineering Journal for Advanced Research

2025 Volume 4 Number 5 September
Publisherwww.singhpublication.com

Infusing Generative AI into Supply Chain Management: Driving Intelligent and Anticipatory Operations

Rao V.V.1*
DOI:10.5281/zenodo.17227221

1* Vijayendra Vittal Rao, Chief Architect (Independent Researcher), IBM, Frisco/Dallas, USA.

Because global supply chains are changing so quickly, businesses need to use new technologies to make them more flexible, accurate, and strong. This study looked into the possible use of Generative Artificial Intelligence (GAI) in supply chain management (SCM) to make operations smarter and more proactive. The study looked at GAI's effects on important supply chain tasks like demand forecasting, inventory optimization, supplier risk assessment, and logistics coordination using a mix of approaches, including simulation modeling and expert reviews. The results showed that GAI made forecasts much more accurate, sped up decision-making, improved inventory levels, and made lead time less variable. Experts agreed that GAI had a lot of technological potential, but they also stressed the importance of human control, interpretability, and ethical use. The results showed how GAI can change the way supply chain ecosystems work by making them more proactive and data-driven, which lets them adapt to changing market conditions.

Keywords: generative AI, supply chain management, demand forecasting, inventory optimization, predictive analytics, intelligent operations, anticipatory decision-making, logistics efficiency, artificial intelligence in SCM

Corresponding Author How to Cite this Article To Browse
Vijayendra Vittal Rao, Chief Architect (Independent Researcher), IBM, Frisco/Dallas, USA.
Email: 		Rao V.V., Infusing Generative AI into Supply Chain Management: Driving Intelligent and Anticipatory Operations. Appl Sci Eng J Adv Res. 2025;4(5):12-18.
Available From
https://asejar.singhpublication.com/index.php/ojs/article/view/165

Manuscript Received Review Round 1 Review Round 2 Review Round 3 Accepted
2025-08-04 2025-08-22 2025-09-10
Conflict of Interest Funding Ethical Approval Plagiarism X-checker Note
None Nil Yes 3.84

© 2025 by Rao VV and Published by Singh Publication. This is an Open Access article licensed under a Creative Commons Attribution 4.0 International License https://creativecommons.org/licenses/by/4.0/ unported [CC BY 4.0].

Appl Sci Eng J Adv Res 2025;4(5)12
Download PDFBack To Article1. Introduction2. Literature
Review3. Result and
Discussion4. ConclusionReferences
Rao V.V. Infusing Generative AI into Supply Chain Management

1. Introduction

Supply chain management (SCM) has changed from a cost-focused task to a strategic driver of business resilience and competitive advantage in today's fast-paced and extremely unstable global market. Traditional supply chain management (SCM) models have had a hard time adapting to changing market conditions, unexpected shocks, and rising consumer demands because they depended too much on historical data and linear forecasting methods. As companies look for ways to make their operations smarter, faster, and more flexible, it is necessary to use new technologies. Generative Artificial Intelligence (GAI) is one of these technologies that has the potential to change the way supply chains work.

Generative AI goes beyond automation to make smart, self-evolving systems. It can develop, simulate, and optimize content and decisions from large data sets. When used in supply chain management (SCM), GAI can make predictions in real time, model supply-demand scenarios, suggest the best logistics routes, and even come up with adaptive techniques to deal with new hazards. Most AI models are merely predictive, while GAI adds a generative layer that not only predicts future states but also offers proactive steps to avoid problems and take advantage of opportunities.

This study looked at how GAI might hypothetically be used in several supply chain activities to see how well it could make operations smarter and more proactive. The study's main goal was to find out how GAI affected important performance areas like anticipating demand, managing inventory, analyzing supplier risk, and coordinating logistics. The study used both simulation modeling and expert validation to try to provide a full picture of how GAI may make supply chains more responsive, efficient, and resilient in a business world that is becoming more complicated.

2. Literature Review

Jackson et al. (2024) suggested a capability-based approach for looking at and using generative AI in supply chain and operations management. Their research focused on putting GAI capabilities into three groups: decision augmentation, autonomous generation, and adaptive learning.

The authors found that GAI might have a big effect on SCM in a few key areas, such as responding to disruptions in real time, controlling inventory in real time, and using logistics systems that optimize themselves. The study also talked about problems with putting the plan into action, like making sure the data fit together, the model was easy to understand, and it fit with the company's overall goal.

Mohamed (2023) looked at the bigger picture of how GAI had changed the way supply chains worked and how well they worked. He said that GAI made supply chains more responsive by making simulations for changes in supply and demand. He also said that its integration let companies switch from reactive to anticipatory supply chain models. The study showed that GAI's unique value proposition is its usage of natural language generation (NLG) for real-time reporting and scenario planning.

Yandrapalli (2023) focused on how GAI would change supply networks, especially how it will make them more visible and flexible. His findings showed that GAI-enabled systems could autonomously generate procurement strategies, supplier risk assessments, and dynamic route optimizations. The study gave examples from the retail and manufacturing sectors where GAI helped speed up operations and improve warehouse management by making smart suggestions.

Hemachandran et al. (2024) looked at how predictive analytics and generative AI may work together in data-driven initiatives, but their main focus was on marketing. The editors did say, though, that the basic ideas of SCM, like adaptive learning, synthetic data generation, and real-time insight production, were also true for SCM. Their talk about how AI can turn unstructured data into useful information was directly related to SCM decision-making settings.

George (2024) looked into how AI may be used in smart manufacturing systems and how it can improve efficiency, quality control, and operational insights. The study wasn't just about the supply chain, but it did give important information about how GAI may be used at the production level to make work instructions, guess when maintenance will be needed, and change production schedules on its own. When these new ideas were used throughout the supply chain, they promised huge increases in efficiency.


Appl Sci Eng J Adv Res 2025;4(5)13
Rao V.V. Infusing Generative AI into Supply Chain Management

Research Methodology

The use of Generative Artificial Intelligence (GAI) in supply chain management (SCM) has changed the way businesses think about operational efficiency and making strategic decisions. Deterministic forecasting and linear workflows were very important to traditional supply chain models, but they didn't always function well in contexts that were unstable or complicated. GAI, on the other hand, might study patterns, make simulations, and come up with adaptive solutions on its own based on real-time data.

This study looked into how GAI could have been used to improve anticipatory skills in SCM tasks like planning for demand, managing inventory, managing supplier risk, and coordinating logistics. The study's goal was to find out if GAI-based systems could be more flexible, accurate, and responsive than traditional ones. The study used simulation models, expert feedback, and key performance indicators (KPIs) to test the idea that GAI may help create smart and proactive supply chain ecosystems.

2.1 Research Design

The study used a mixed-methods exploratory research design, which means it used both quantitative and qualitative methods. We used simulation modeling to create supply chain scenarios with and without GAI in a quantitative way. We did expert interviews and validation sessions to look at the perceived impact, practicality, and trustworthiness of GAI outputs in supply chain decision-making.

This research design made it possible to do a full analysis by balancing the accuracy of computers with the knowledge of people. The mixed-method approach made sure that the study looked at both the technical efficiency and strategic importance of adding GAI to SCM operations.

2.2 Data Collection

Primary Data

We got primary data by doing semi-structured interviews with 15 professionals, such as supply chain managers, AI engineers, logistics analysts, and data scientists. The goal of these interviews was to get real-world opinions on what GAI can and can't do in operational settings.

There were also expert validation sessions where participants were shown GAI-generated outputs such risk maps, forecast demand projections, and inventory policies. We used their feedback to find out how easy it was to understand and use the outcomes in real supply chain situations.

Secondary Data

We got our secondary data from publicly available datasets like Kaggle supply chain repositories, synthetic ERP databases, and case studies that were specific to the industry. These datasets gave us information about past demand patterns, shipment delays, stock levels, and how well suppliers were doing. We also examined literature reviews and white papers from big consulting companies and AI solution vendors to put the results in context.

2.3 Generative AI Framework Simulation

A GAI simulation framework was developed using tools such as OpenAI’s GPT architecture, Python, PyTorch, and relevant transformer models. The framework was designed to mimic real-time data environments and autonomously generate decisions and simulations.

The GAI model was configured to:

  • Generate accurate demand forecasts using historical sales, seasonal trends, and macroeconomic indicators.
  • Recommend optimal inventory strategies using reinforcement learning to reduce holding costs and avoid stockouts.
  • Simulate supplier behavior under scenarios like raw material shortages or geopolitical instability.
  • Produce natural language summaries and dashboards to assist in managerial decision-making.

These simulations allowed for a controlled evaluation of GAI’s performance under both stable and disruptive supply chain conditions.

2.4 Evaluation Metrics

To evaluate the performance of GAI-infused SCM, the study utilized a set of standard KPIs:

  • Forecast Accuracy: Measured using Mean Absolute Percentage Error (MAPE) and Root Mean Square Error (RMSE).

Appl Sci Eng J Adv Res 2025;4(5)14
Rao V.V. Infusing Generative AI into Supply Chain Management

  • Inventory Turnover Ratio: Assessed improvements in stock utilization and minimization of dead stock.
  • Lead Time Variability: Measured reduction in variability and delays across the supply network.
  • Decision Latency: Evaluated the time taken to process data and produce actionable recommendations using GAI.

Each KPI was calculated for both pre-GAI and post-GAI simulations to enable comparative analysis.

2.5 Analytical Methods

The study utilized descriptive statistics to look at raw data and inferential statistics like t-tests and ANOVA to find big changes in performance.

We used agent-based modeling to create scenario-based simulations that showed what would happen if a supply failed or demand suddenly rose. We used thematic analysis on the transcripts of the interviews to find the main ideas about trust, how easy it is to understand, and how ready people are for AI integration.

All results were triangulated to ensure validity and consistency across data sources and analytical techniques.

3. Result and Discussion

This part talks about the results of putting Generative AI (GAI) into different parts of supply chain activities in a simulated way. The study looked at the effectiveness of GAI in allowing anticipatory and intelligent decision-making in demand forecasting, inventory management, supplier risk evaluation, and logistics planning by looking at key performance indicators (KPIs), expert feedback, and simulation modeling. We looked at the data to see both the quantitative improvements and the qualitative insights that could help us grasp the strategic benefits and practical constraints of using GAI.

3.1 Improvement in Forecast Accuracy

The GAI-enhanced demand forecasting model was far more accurate than classic statistical models like ARIMA and moving averages. The GAI model was better at capturing complicated demand patterns that were affected by both internal and external factors since the Mean Absolute Percentage Error (MAPE) and Root Mean Square Error (RMSE) went down.

Table 1: Forecasting Accuracy Comparison

Model UsedMAPE (%)RMSE (Units)
Traditional (ARIMA)12.8420
GAI Model5.4210

asejar_165_01.PNG
Figure 1: Forecasting Accuracy Comparison

When comparing the accuracy of predictions made by traditional ARIMA models and the Generative AI (GAI) model, it was clear that GAI made a big difference in performance. The Mean Absolute Percentage Error (MAPE) went down from 12.8% to 5.4%, which means that demand trends may be predicted more accurately. The Root Mean Square Error (RMSE) also went down from 420 units to 210 units, which shows that the size of the forecast errors got a lot smaller. These results showed that the GAI model was better at capturing complex patterns and variables that affect demand. This made supply chain planning more accurate and dependable.

3.2 Optimization in Inventory Management

GAI-generated inventory policies led to improved inventory turnover ratios and reduced excess stock. The model suggested dynamic reorder points and safety stock levels based on real-time demand simulations and risk profiles.

Table 2: Inventory Performance Metrics

MetricPre-GAI (Baseline)Post-GAI
(With Generative AI)
Inventory Turnover Ratio4.26.8
Average Stock-Out Events/Year186
Holding Cost (USD/year)$7500$4750

Comparing inventory performance measures before and after the introduction of Generative measures before and after the introduction of Generative AI (GAI) showed significant improvements in how well inventory was managed and how much it cost.


Appl Sci Eng J Adv Res 2025;4(5)15
Rao V.V. Infusing Generative AI into Supply Chain Management

The inventory turnover ratio went up from 4.2 to 6.8, which means that inventory was used more efficiently and moved more quickly. Also, the average number of stock-outs each year went down a lot, from 18 to 6. This shows that demand forecasting and replenishment accuracy have gotten better. Holding costs also went down from $7500 to $4750 a year, which shows that GAI can optimize stock levels and cut down on extra inventory. Overall, these changes showed that GAI helped make inventory management more efficient and responsive, which led to better service and lower costs.

3.3 Lead Time and Logistics Efficiency

GAI contributed to better logistics coordination by forecasting potential disruptions (e.g., weather, traffic, supplier delays) and recommending rerouting or alternative scheduling. The lead time variability was significantly minimized.

Table 3: Logistics Efficiency Indicators

ParameterWithout GAIWith GAI
Average Lead Time (days)6.44.8
Lead Time Variability (%)28.511.3
Late Deliveries (%)22.18.4

asejar_165_02.PNG
Figure 2: Logistics Efficiency Indicators

Comparing logistics performance metrics before and after adding Generative AI (GAI) showed a big increase in the dependability and efficiency of the supply chain. The average lead time went down from 6.4 days to 4.8 days, which shows that GAI can make operations more efficient and predict delays. More importantly, lead time variability went down from 28.5% to 11.3%, which shows that delivery schedules are now much more predictable and consistent. Also, the number of late deliveries went down from 22.1% to 8.4%, which shows that GAI can proactively manage risks and make the best routing decisions.

These changes showed how GAI helped make the supply chain system more flexible and reliable.

3.4 Decision-Making Latency Reduction

GAI systems automated the generation of dashboards and actionable insights in near real-time. This drastically reduced decision latency, empowering managers to act swiftly on supply chain disruptions.

Table 4: Decision-Making Latency

Decision TypeAverage Time (Manual)Time with GAI Support
Reorder Triggering3 hours20 minutes
Supplier Reallocation5 hours40 minutes
Route Optimization4 hours30 minutes

Comparing the time it takes to make decisions between manual operations and those that use Generative AI (GAI) showed that operational latency was much reduced in crucial supply chain tasks. Specifically, the time it took to trigger a reorder went from 3 hours to only 20 minutes. This shows that GAI can quickly examine inventory data and make decisions about whether to restock. Also, supplier reallocation, which used to take up to five hours because someone had to manually check on vendor performance and availability, only took 40 minutes with GAI's help. Route optimization went from taking 4 hours to 30 minutes, showing how well GAI can handle complicated logistics factors and come up with the best delivery routes in almost real time. These results showed that GAI greatly sped up decision-making, which made supply chain operations more flexible and responsive.

3.5 Qualitative Insights from Expert Interviews

Experts who reviewed the GAI-generated simulations reported increased confidence in AI-assisted planning. However, they expressed concerns regarding model interpretability and the need for human oversight in high-risk decisions (e.g., supplier blacklisting or high-value order adjustments).

Key themes from qualitative analysis included:

  • Trust in AI decisions improved when transparent rationale was provided.
  • Data governance and ethical use of AI remained top organizational concerns.

Appl Sci Eng J Adv Res 2025;4(5)16
Rao V.V. Infusing Generative AI into Supply Chain Management

  • Upskilling needs were identified, especially in interpreting AI-generated analytics and simulations.

Discussion

The results showed that GAI might be a key factor in changing how supply chains work by making predictions more accurate and allowing for proactive actions. The technology's ability to reduce common SCM bottlenecks was shown by the measurable improvements in forecasting, inventory control, logistics efficiency, and decision time.

But the results also showed how important it is to have human-in-the-loop frameworks to make sure that things are clear, flexible, and ethically overseen. To make it work, companies would have to spend money on digital infrastructure, cross-functional training, and ways to handle change.

4. Conclusion

The study above found that using Generative AI in supply chain management greatly improved operational performance by making demand forecasting more accurate, optimizing inventory levels, reducing lead time variability, and speeding up decision-making. The simulation findings showed that essential supply chain functions became much more efficient, responsive, and cost-effective. Experts also agreed that GAI had strategic potential, but they also stressed the importance of keeping human oversight, assuring transparency, and resolving ethical issues. In general, GAI made it possible for supply chain operations to be smart and forward-thinking in business settings that are dynamic and complicated.

References

1. A. S. George. (2024). AI-enabled intelligent manufacturing: A path to increased productivity, quality, and insights. Partners Univ. Innov. Res. Publ., 2(4), 50–63.

2. A. Sharma. (2024). From blueprint to flight: Guiding your first generative AI project—revolutionizing service desk operations.

3. D. Brunner, C. Legat, & U. Seebacher. (2024). Towards next generation data-driven management: Leveraging predictive swarm intelligence to reason and predict market dynamics. in Collective Intelligence, CRC Press, pp. 152–203.

4. D. Brunner, C. Legat, & U. Seebacher. (2024). Towards next generation data-driven management: Collective intelligence—The rise of swarm systems and their impact on society. CRC Press, p. 152.

5. F. Dou et al. (2023). Towards artificial general intelligence (AGI) in the internet of things (IoT): Opportunities and challenges. arXiv:2309.07438.

6. F. Jia, S. Hu, Q. He, L. Chen, & X. Xing. The role of artificial intelligence in supply chain risk management: Towards an integrated conceptual framework. Available at: SSRN 4913631.

7. I. Jackson, D. Ivanov, A. Dolgui, & J. Namdar. (2024). Generative artificial intelligence in supply chain and operations management: a capability-based framework for analysis and implementation. Int. J. Prod. Res., 62(17), 6120–6145.

8. K. Hemachandran, D. Choudhury, R. V. Rodriguez, J. A. Wise, & T. Revathi. (2024). Predictive analytics and generative AI for data-driven marketing strategies. Boca Raton, FL, USA: CRC Press.

9. L. Luo. (2024). A web intelligent hotel management framework based on IoT and generative AI. J. Web Eng., 23(7), 885–912.

10. M. Kmiecik. (2023). ChatGPT in third-party logistics–The game-changer or a step into the unknown?, J. Open Innov. Technol. Mark. Complex., 9(4), 100174.

11. O. A. M. Mohamed. (2023). How generative AI transforming supply chain operations and efficiency?.

12. R. L. Julie, S. Raja, P. T. Selvan, M. R. Priya, & N. K. Rajagopal. (2024). Exploring the transformative effects of AI on entrepreneurship in business performance. in Balancing Automation and Human Interaction in Modern Marketing, IGI Global, pp. 127–150.

13. V. Kumar, & P. Kotler. (2024). A market management approach to transformative business operations. Mark. Strategy J., 1, 100005.

14. V. Yandrapalli. (2023). Revolutionizing supply chains using power of generative AI. Int. J. Res. Publ. Rev., 4(12), 1556–1562.


Appl Sci Eng J Adv Res 2025;4(5)17
Rao V.V. Infusing Generative AI into Supply Chain Management

15. W. Lo, C. M. Yang, Q. Zhang, & M. Li. (2024). Increased productivity and reduced waste with robotic process automation and generative AI-powered IoE services. J. Web Eng., 23(1), 53–87.

Disclaimer / Publisher's Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of Journals and/or the editor(s). Journals and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

Appl Sci Eng J Adv Res 2025;4(5)18