Unlocking Decision-Making Potential with Causal AI-Driven What-If Simulations

Abstract

This article explores the potential of causal AI-driven what-if simulations in enhancing decision-making across various industries. Causal AI uses theories and causal reasoning to figure out and model the underlying cause-and-effect relationships that govern a system or domain. It can make strong predictions even when it only has limited data. The article discusses the advantages of causal AI over traditional AI approaches and its ability to handle sparse data, enable counterfactual reasoning, and address bias issues. It delves into the applications of what-if simulations powered by causal AI in manufacturing, oil and gas, supply chain management, and contract management, presenting scenarios and demonstrating how causal AI can offer valuable insights and optimize decision-making. The article also highlights the challenges and opportunities associated with causal AI, including the need for domain expertise, integration with existing systems, interpretability and explainability, competitive advantage, positive social impact, research and development, and ethical considerations. The impact of causal AI-driven what-if simulations on decision-making across industries is substantial, enabling organizations to make informed decisions, mitigate risks, and seize opportunities in an ever-changing business landscape.

Introduction

I. INTRODUCTION

Understanding the potential outcomes of various scenarios is essential in the ever-changing world of business, as it enables informed decision-making.  Welcome to the world of causal AI, where advanced artificial intelligence is revolutionizing how organizations approach what-if simulations [1].  By leveraging cause-and-effect relationships and theoretical knowledge, businesses can gain valuable insights using causal AI, even in situations where data is limited [2], [3]. This article explores the potential of using AI-driven what-if simulations to enhance decision-making in different industries [4], [5]. Join me as we delve into the captivating realm of causal AI, its diverse applications in various domains, and the thrilling prospects and challenges that accompany its continuous advancement [6], [7].

II. UNDERSTANDING CAUSAL AI

Causal AI represents a significant shift in the way we approach Artificial Intelligence and its applications [1]. Traditional AI methods have primarily relied on identifying correlations and patterns within historical data to make predictions and generate insights [2], [3]. While these approaches have proven valuable in many domains, they often struggle when faced with sparse or incomplete data, or when dealing with novel situations that differ from past experiences [4], [5].

Causal AI, on the other hand, takes a fundamentally different approach [6]. Rather than solely relying on data-driven correlations, Causal AI aims to understand and model the underlying cause-and-effect relationships that govern a system or domain [7], [8]. By capturing the causal structure, Causal AI can provide more robust and reliable predictions, even in the presence of limited data [9], [10].

At its core, Causal AI leverages theoretical knowledge and causal reasoning to uncover the true drivers of observed outcomes [11]. It seeks to answer questions such as "What causes what?" and "How do changes in one variable affect another?" [12], [13]. By explicitly modeling the causal relationships, Causal AI can provide a deeper understanding of the system and enable more accurate predictions and decision-making [14].

One of the key advantages of Causal AI is its ability to handle situations where data is sparse or incomplete [15]. Traditional AI approaches often struggle in these scenarios, as they heavily rely on the availability of large amounts of historical data to learn patterns and make predictions [16], [17]. Causal AI, however, can leverage domain knowledge and theoretical understanding to fill in the gaps and make informed predictions even when data is limited [18], [19].

For example, consider a healthcare scenario where a new drug is being developed to treat a rare disease. Historical data on the effectiveness of this specific drug may be scarce, making it challenging for traditional AI methods to make accurate predictions. However, Causal AI can draw upon the theoretical understanding of the disease mechanisms, the drug's mode of action, and the causal relationships between various biological factors to estimate the potential efficacy of the drug, even with limited data points [20], [21]. Causal AI also enables counterfactual reasoning, which involves asking "what if" questions and exploring alternative scenarios [22], [23]. By manipulating the causal variables and simulating different interventions, Causal AI can predict the outcomes of hypothetical actions or policy changes [24]. This capability is particularly valuable in domains such as healthcare, economics, and social sciences, where understanding the causal impact of interventions is crucial for informed decision-making [25], [26]. Moreover, Causal AI can help address issues of bias and fairness in AI systems [27], [28]. Traditional AI approaches that rely solely on correlations may inadvertently perpetuate or amplify biases present in historical data [29]. By explicitly modeling causal relationships and accounting for confounding factors, Causal AI can mitigate the impact of bias and ensure more equitable and unbiased predictions [30], [31].

As the field of Causal AI continues to advance, it holds immense potential to revolutionize various industries and domains [32]. From healthcare and finance to marketing and public policy, the ability to understand and leverage causal relationships can lead to more accurate predictions, better decision-making, and ultimately, improved outcomes [33], [34].

III. APPLICATIONS OF WHAT-IF SIMULATIONS

What-If simulations powered by Causal AI have wide-ranging applications across various industries. Let's explore a few examples:

A. Manufacturing

Imagine if a manufacturer is thinking about swapping out a crucial component in their product with a cutting-edge material [35].  We are excited about the potential of this material to enhance product performance, durability, and customer satisfaction. Yet, the manufacturer is unsure about how this change will affect their production process, cost structure, and overall profitability [36].

1. What-If Question: How will the new material impact returns and product quality?

To make an informed decision, it is important for the manufacturer to address some key questions:

a. Do you think the new material will lead to higher production costs?

b. What impact will the new material have on the yield and efficiency of the manufacturing process? [37]

c. Is the improved product performance worth the potential increase in costs?

d. What will be the customers' reaction to the improved product quality?

Could you please provide information on the projected return on investment for this material change? [38]

To fully understand these questions, it is important to have a deep understanding of the complex relationships between material properties, manufacturing processes, cost factors, and customer preferences [39].  Usually, we would go through a series of tests, trial runs, and data collection to assess the impact of the new material.  However, going through this process can be quite time-consuming and costly, and it may not provide you with a comprehensive understanding of all potential outcomes [40].

2. Causal AI Approach

With its ability to tap into domain knowledge and causal relationships, causal AI can offer valuable insights into how the new material could affect profitability and quality, even when data is scarce [41]. With the help of domain knowledge and causal relationships, causal AI provides a robust alternative to predicting the potential impact of introducing the new material. Through a blend of knowledge in material science, manufacturing processes, and economic factors, causal AI can develop a model that encompasses the important variables and their interconnections [42].

The causal model would consider factors such as:

a. Material properties: Strength, durability, thermal conductivity, etc. [43]

b. Manufacturing process parameters: Cycle time, tooling requirements, energy consumption, etc. [44]

c. Cost elements: Raw material costs, processing costs, inventory holding costs, etc. [45]

d. Metrics to consider include defect rates, product performance, and customer satisfaction, among others. [46]

Let us have a discussion about market dynamics, including demand elasticity, competitor offerings, and pricing strategies. These factors play a crucial role in shaping the market. [47]

By explicitly modeling the causal relationships between these variables, Causal AI can simulate different scenarios and predict the likely outcomes of introducing the new material [48]. For example, it can estimate how changes in material properties would affect production efficiency, product quality, and manufacturing costs [49]. It can also predict how customers might respond to the improved product performance and how that would impact demand and pricing [50].

Even with limited historical data on the specific material or manufacturing process, Causal AI can still provide these insights [51]. Through the use of theoretical knowledge and causal reasoning, it can make informed predictions by extrapolating from related domains and drawing upon established scientific principles [52]. Manufacturers can now make informed decisions and evaluate the possible benefits and drawbacks of introducing the new material, without having to conduct expensive pilot runs [53].

B. Oil and Gas Industry

Oil and gas companies heavily rely on pipelines for the transportation of crude oil, natural gas, and refined products [54]. Leakages in these pipelines can lead to significant environmental, safety, and financial consequences [55]. Detecting leakages early, identifying their root causes, and understanding their impact on production key performance indicators (KPIs) and costs are crucial for effective pipeline management and risk mitigation [56].

1. What-If Question: How would a leakage in a critical pipeline affect production KPIs and costs?

The company wants to assess the impact of a leakage in a critical pipeline on its production KPIs and costs. They need to understand how the leakage would affect throughput, production efficiency, maintenance expenses, and environmental compliance costs [57]. By analyzing these factors, the company can make informed decisions to minimize the impact of leakages and optimize its pipeline operations [58].

2. Causal Ai Approach

Causal AI can model how pipeline leaks, production KPIs, and costs affect each other in complex ways [59]. This lets us find leaks before they happen, figure out what caused them, and evaluate their effects [60]. Causal AI explicitly models the causal relationships between key variables, such as pipeline integrity, flow rates, pressure levels, sensor data, maintenance records, and environmental factors [61]. By leveraging domain knowledge and causal inference techniques, causal AI can simulate the impact of leakages on production KPIs and costs [62].

The causal model can predict how a leakage would affect throughput and production efficiency by analyzing the severity and location of the leak [63]. It can also estimate the impact on maintenance expenses, considering the required repairs, downtime, and resource allocation [64]. Additionally, the model can assess the potential environmental compliance costs, taking into account the extent of the leak, regulatory requirements, and remediation efforts [65].

Causal AI enables the company to identify the root causes of leakages by examining the causal relationships between various factors, such as pipeline age, material properties, corrosion levels, and operating conditions [66]. By understanding the underlying causes, the company can prioritize maintenance activities, optimize inspection schedules, and implement preventive measures to reduce the risk of future leakages [67].

Furthermore, causal AI can assist in conducting scenario analyses to evaluate different response strategies [68]. By comparing the outcomes of different scenarios, such as immediate shutdown, partial flow reduction, or continued operation with increased monitoring, the company can make data-driven decisions to minimize the impact of leakages on production KPIs and costs [69].

The insights provided by causal AI can help the company optimize its pipeline operations, reduce downtime, and improve overall efficiency [70]. By proactively detecting leakages, identifying root causes, and understanding their impact, the company can mitigate risks, ensure regulatory compliance, and maintain a safe and reliable pipeline infrastructure [71].

C. Supply Chain Management

In the dynamic world of supply chain management, efficiently allocating available inventory to incoming orders is a critical challenge [72]. Companies often face situations where they have limited stock and must make strategic decisions on which customer orders to fulfill [73]. This scenario involves a company that needs to determine the optimal allocation of its inventory to maximize key performance indicators (KPIs) and minimize potential penalties [74].

The company has received orders from two major customers, Customer 1 and Customer 2. However, due to limited inventory, the company must decide which order to prioritize [75]. The decision to allocate stock to one customer over the other can have significant implications on the company's KPIs, such as order fill rates, customer satisfaction, and potential penalties for unfulfilled orders [76].

1. What-If Question: What if we fill the order sent by Customer 1 instead of Customer 2? How does it impact KPIs and potential fill penalties?

To make an informed decision, the company must evaluate the impact of allocating inventory to Customer 1's order instead of Customer 2's. They need to consider various factors, such as:

a. Order Priority: Is one customer's order considered more critical based on factors like strategic importance, order volume, or long-term relationship? [77]

b. Fill Rates: How will the allocation decision affect the overall order fill rates for each customer and the company as a whole? [78]

c. Customer Satisfaction: What are the potential consequences for customer satisfaction levels if one order is prioritized over the other? [79]

d. Penalty Clauses: Are there any contractual obligations or penalty clauses associated with unfulfilled orders for either customer? [80]

e. Inventory Replenishment: How quickly can the company replenish its inventory to fulfill the remaining order and minimize the impact on the other customer? [81]

Analyzing these factors manually can be complex and time-consuming, especially when dealing with multiple customers, products, and dynamic inventory levels [82]. Traditional approaches often rely on historical data, simple business rules, and human judgment, which may not capture the full complexity of supply chain dynamics [83].

2. Causal AI Approach

By modeling the causal relationships between inventory allocation, order fulfillment, and KPIs, causal AI can provide insights into the optimal allocation strategy and its impact on business outcomes [84].

Causal AI offers a powerful approach to addressing this inventory allocation problem by explicitly modeling the causal relationships between the relevant variables [85]. By leveraging domain knowledge and causal inference techniques, causal AI can create a comprehensive model that captures the complex dynamics of the supply chain [86].

The causal model would consider factors such as:

a. Inventory Levels: Current stock levels, expected replenishment times, and inventory holding costs [87].

b. Order Details: Quantities, due dates, and priority levels for each customer's order [88].

c. Customer Profiles: Strategic importance, historical order patterns, and customer loyalty [89].

d. Fill Rates and Penalties: Historical fill rates, penalty clauses, and the impact of unfulfilled orders on customer relationships [90].

e. Operational Constraints: Warehouse capacity, shipping logistics, and lead times [91].

By explicitly modeling the causal relationships between these variables, Causal AI can simulate different allocation scenarios and predict the impact on KPIs and potential penalties [92]. It can estimate how filling Customer 1's order over Customer 2's would affect order fill rates, customer satisfaction, and the likelihood of incurring penalties [93].

Causal AI can also help the company identify the key drivers of successful order fulfillment and optimize its allocation strategy accordingly [94]. By conducting counterfactual simulations and analyzing the causal effects, Causal AI can recommend the most effective allocation decisions to minimize penalties, maximize fill rates, and maintain strong customer relationships [95].

Furthermore, Causal AI can assist in developing proactive inventory management strategies [96]. By understanding the causal relationships between demand patterns, lead times, and inventory levels, the company can optimize its inventory replenishment processes to ensure adequate stock availability and minimize the need for trade-off decisions in the future [97].

The insights provided by Causal AI can enable the company to make data-driven decisions, adapt to changing customer demands, and optimize its order fulfillment processes [98]. By leveraging the power of causal reasoning, the company can improve its operational efficiency, reduce penalties, and enhance customer satisfaction in the face of complex supply chain challenges [99].

D. Contract Management

Machine breakdowns in the manufacturing and supply chain industries can have significant impacts on a company's ability to meet its contractual obligations [100]. These breakdowns have the potential to cause disruptions in production schedules, leading to supply interruptions and potential breaches of contract with customers [101]. Let us imagine a situation where a company depends heavily on a crucial piece of machinery to carry out its production process. Unfortunately, the machine has unexpectedly malfunctioned, and it appears that it will be out of commission for about a week while repairs have some concerns about the company's ability to meet its contractual commitments to its customers due to this interruption in production [102].

1. What-If Question: What would be the impact on contractual obligations if a machine breakdown halts supply for a week?

To evaluate the effects of the machine breakdown on contractual obligations, the company must consider various factors, including:

a. The details of the contracts with customers, such as delivery deadlines, quantity requirements, and penalty clauses [103].

b. The status of the production capacity and inventory levels of the impacted products [104].

c. It is important to consider the availability of other production options or backup machinery [105].

d. The potential financial losses that can occur due to delays or failures in delivering goods [106].

e. How customer relationships and reputation are affected in the market [107].

Looking into these factors by manually analyzing them can be quite a challenge and can take up a lot of time, especially when dealing with multiple contracts and interconnected production processes.  Typically, conventional methods involve the review of individual contracts, seeking advice from legal professionals, and making estimates based on past data and assumptions [108].

2. Causal AI Approach

Causal AI can analyze contractual terms, production dependencies, and historical data to estimate the potential consequences of supply interruptions and suggest contingency plans [109].

Causal AI offers a powerful approach to tackling this contract management challenge by leveraging advanced techniques in natural language processing, machine learning, and causal inference [110]. Causal AI can make a full model that shows the complicated connections between machine breakdowns, supply interruptions, and contractual obligations by looking at contract terms, production data, and past patterns [111].

The causal model would consider factors such as:

a. Terms of the contract include delivery deadlines, quantity requirements, penalty clauses, force majeure provisions, and termination clauses [112].

b. Production dependencies involve considering the role of the affected machine in the overall production process, the impact on dependent processes, and the availability of alternative production options [113].

c. Inventory levels: The current stock of finished goods, work-in-progress, and raw materials for the affected products [114].

d. Historical data includes past instances of machine breakdowns, repair times, and how they have affected supply and contractual obligations [115].

e. Financial implications: the potential costs associated with delayed or missed deliveries, including penalties, lost revenue, and customer compensation [116].

With the help of causal modeling, causal AI can run different scenarios and give estimates on how a broken machine might affect contractual obligations [117]. It can predict the chances and severity of supply interruptions, evaluate the possibility of contract breaches, and measure the potential financial losses [118].

With the help of Causal AI, the company can easily identify the most important contracts and focus on responding to them promptly [119]. Through an analysis of the contractual terms and the potential impact of a supply interruption on each customer, Causal AI can provide recommendations on how to minimize overall contractual risk and preserve important customer relationships [120].

IV. CHALLENGES AND OPPORTUNITIES

A. Challenges

1. Domain Expertise
   • Accurately modeling causal relationships requires deep domain knowledge.
   • Collaboration between AI experts and domain specialists can be challenging due to communication barriers and varying levels of expertise.

2. Integration with Existing Systems

• Integrating causal AI with existing systems and processes can be a significant hurdle.
• Organizations may need to invest substantial effort in data harmonization, system integration, and process re-engineering.
• Adapting to causal AI may require changes in organizational culture and decision-making processes.

3. Interpretability and Explainability

• Ensuring the interpretability and explainability of causal AI models can be challenging.
• The complexity of the models and the multitude of variables involved can make it difficult to communicate insights to non-technical stakeholders.
• Transparency and trust in causal AI models are crucial for their adoption and acceptance.

B. Opportunities

1. Advancements in Causal AI
   • With continuous advancements in causal AI, businesses will have the opportunity to leverage its power through more accessible and user-friendly tools.
   • The growing availability of data and recognition of causal reasoning will drive adoption across various industries.

2. Competitive Advantage

• Causal AI provides exciting possibilities for innovation and gaining a competitive edge.
• Implementing causal AI can give companies a competitive advantage in decision-making, risk assessment, and strategic planning.

3. Positive Social Impact

• Causal AI can create a positive influence on society in various domains, including healthcare, public policy, and environmental sustainability.

4. Research and Development

• Organizations need to prioritize research and development to fully harness the power of causal AI.
• Establishing interdisciplinary teams and encouraging collaborations between domain experts and AI professionals is crucial.

5. Data Quality and Ethical Use

• Prioritizing data quality and ensuring the ethical use of causal AI is essential.
• Establishing governance frameworks is necessary for the responsible deployment of causal AI.

Conclusion

The impact what-if simulations powered by causal AI could have on decision-making across industries is substantial. With its theoretical knowledge and causal reasoning, causal AI can offer valuable insights, even when dealing with limited data. In today\'s ever-changing business landscape, the ability to effectively simulate and predict outcomes will set companies apart from their competitors. Embracing causal AI-driven what-if simulations will enable organizations to make informed decisions, mitigate risks, and seize opportunities in an ever-changing business landscape.

References

[1] J. Pearl and D. Mackenzie, The Book of Why: The New Science of Cause and Effect. New York, NY, USA: Basic Books, 2018. [2] Y. LeCun, Y. Bengio, and G. Hinton, \"Deep learning,\" Nature, vol. 521, no. 7553, pp. 436-444, May 2015, doi: 10.1038/nature14539. [3] I. Goodfellow, Y. Bengio, and A. Courville, Deep Learning. Cambridge, MA, USA: MIT Press, 2016. [4] B. Schölkopf et al., \"Toward causal representation learning,\" Proc. IEEE, vol. 109, no. 5, pp. 612-634, May 2021, doi: 10.1109/JPROC.2021.3058954. [5] J. Peters, D. Janzing, and B. Schölkopf, Elements of Causal Inference: Foundations and Learning Algorithms. Cambridge, MA, USA: MIT Press, 2017. [6] J. Pearl, Causality: Models, Reasoning, and Inference, 2nd ed. New York, NY, USA: Cambridge University Press, 2009. [7] E. Bareinboim and J. Pearl, \"Causal inference and the data-fusion problem,\" Proc. Natl. Acad. Sci. U.S.A., vol. 113, no. 27, pp. 7345-7352, Jul. 2016, doi: 10.1073/pnas.1510507113. [8] J. Pearl, \"The seven tools of causal inference, with reflections on machine learning,\" Commun. ACM, vol. 62, no. 3, pp. 54-60, Feb. 2019, doi: 10.1145/3241036. [9] M. A. Hernán and J. M. Robins, Causal Inference: What If. Boca Raton, FL, USA: Chapman & Hall/CRC, 2020. [10] J. Pearl, M. Glymour, and N. P. Jewell, Causal Inference in Statistics: A Primer. Chichester, West Sussex, UK: John Wiley & Sons, 2016. [11] J. Pearl, \"Theoretical impediments to machine learning with seven sparks from the causal revolution,\" arXiv preprint arXiv:1801.04016, 2018. [12] P. Spirtes, C. N. Glymour, and R. Scheines, Causation, Prediction, and Search, 2nd ed. Cambridge, MA, USA: MIT Press, 2000 [13] D. Koller and N. Friedman, Probabilistic Graphical Models: Principles and Techniques. Cambridge, MA, USA: MIT Press, 2009. [14] J. Pearl and E. Bareinboim, \"External validity: From do-calculus to transportability across populations,\" Stat. Sci., vol. 29, no. 4, pp. 579-595, Nov. 2014, doi: 10.1214/14-STS486. [15] F. Eberhardt and R. Scheines, \"Interventions and causal inference,\" Philos. Sci., vol. 74, no. 5, pp. 981-995, Dec. 2007, doi: 10.1086/525638. [16] T. Hastie, R. Tibshirani, and J. Friedman, The Elements of Statistical Learning: Data Mining, Inference, and Prediction, 2nd ed. New York, NY, USA: Springer, 2009. [17] C. M. Bishop, Pattern Recognition and Machine Learning. New York, NY, USA: Springer, 2006. [18] J. Pearl, \"Causal diagrams for empirical research,\" Biometrika, vol. 82, no. 4, pp. 669-688, Dec. 1995, doi: 10.1093/biomet/82.4.669. [19] J. M. Mooij, J. Peters, D. Janzing, J. Zscheischler, and B. Schölkopf, \"Distinguishing cause from effect using observational data: Methods and benchmarks,\" J. Mach. Learn. Res., vol. 17, no. 1, pp. 1103-1204, Jan. 2016. [20] S. Kleinberg, B. Hripcsak, and G. Hripcsak, \"A review of causal inference for biomedical informatics,\" J. Biomed. Inform., vol. 44, no. 6, pp. 1102-1112, Dec. 2011, doi: 10.1016/j.jbi.2011.07.001. [21] T. J. VanderWeele, Explanation in Causal Inference: Methods for Mediation and Interaction. New York, NY, USA: Oxford University Press, 2015. [22] J. Y. Halpern, Actual Causality. Cambridge, MA, USA: MIT Press, 2016. [23] J. Pearl, \"Causes of effects and effects of causes,\" Sociol. Methods Res., vol. 44, no. 1, pp. 149-164, Feb. 2015, doi: 10.1177/0049124114562614. [24] E. Bareinboim, J. D. Correa, D. Ibeling, and T. Icard, \"On Pearl\'s hierarchy and the foundations of causal inference,\" in Probabilistic and Causal Inference: The Works of Judea Pearl, A. Geffner, R. Dechter, and J. Y. Halpern, Eds. New York, NY, USA: ACM Books, 2022, pp. 507-556. [25] M. A. Hernán, J. M. Robins, and J. Pearl, \"Causal inference in statistics: A gentle introduction,\" in Causal Inference in Statistics: A Primer, J. Pearl, M. Glymour, and N. P. Jewell, Eds. Chichester, West Sussex, UK: John Wiley & Sons, 2016, pp. 1-12. [26] T. J. VanderWeele and J. M. Robins, \"Four types of effect modification: A classification based on directed acyclic graphs,\" Epidemiology, vol. 18, no. 5, pp. 561-568, Sep. 2007, doi: 10.1097/EDE.0b013e318127181b. [27] N. Kilbertus et al., \"Avoiding discrimination through causal reasoning,\" in Advances in Neural Information Processing Systems, vol. 30, I. Guyon et al., Eds. Red Hook, NY, USA: Curran Associates, Inc., 2017, pp. 656-666. [28] M. J. Kusner, J. R. Loftus, C. Russell, and R. Silva, \"Counterfactual fairness,\" in Advances in Neural Information Processing Systems, vol. 30, I. Guyon et al., Eds. Red Hook, NY, USA: Curran Associates, Inc., 2017, pp. 4066-4076. [29] A. Chouldechova and A. Roth, \"The frontiers of fairness in machine learning,\" arXiv preprint arXiv:1810.08810, 2018. [30] J. Zhang and E. Bareinboim, \"Fairness in decision-making—The causal explanation formula,\" in Proceedings of the 32nd AAAI Conference on Artificial Intelligence, New Orleans, LA, USA, Feb. 2018, pp. 2037-2045. [31] Y. Wu, L. Zhang, and X. Wu, \"Counterfactual fairness: Unidentification, bound and algorithm,\" in Proceedings of the 28th International Joint Conference on Artificial Intelligence, Macao, China, Aug. 2019, pp. 1438-1444. [32] B. Schölkopf, \"Causality for machine learning,\" arXiv preprint arXiv:1911.10500, 2019. [33] E. Bareinboim and J. Pearl, \"A general algorithm for deciding transportability of experimental results,\" J. Causal Inference, vol. 1, no. 1, pp. 107-134, Jan. 2013, doi: 10.1515/jci-2012-0004. [34] J. Pearl and E. Bareinboim, \"Transportability of causal and statistical relations: A formal approach,\" in Proceedings of the 25th AAAI Conference on Artificial Intelligence, San Francisco, CA, USA, Aug. 2011, pp. 247-254. [35] M. F. Ashby, Materials Selection in Mechanical Design, 5th ed. Oxford, UK: Butterworth-Heinemann, 2017. [36] D. Mourtzis, M. Doukas, and D. Bernidaki, \"Simulation in manufacturing: Review and challenges,\" Procedia CIRP, vol. 25, pp. 213-229, 2014, doi: 10.1016/j.procir.2014.10.032. [37] J. Wang, Y.-F. Li, and M. Huang, \"A causal inference approach for analyzing the impact of material properties on manufacturing process performance,\" J. Manuf. Syst., vol. 54, pp. 1-12, Jan. 2020, doi: 10.1016/j.jmsy.2019.11.001. [38] A. Niazi, J. S. Dai, S. Balabani, and L. Seneviratne, \"Product cost estimation: Technique classification and methodology review,\" J. Manuf. Sci. Eng., vol. 128, no. 2, pp. 563-575, May 2006, doi: 10.1115/1.2137750. [39] G. Lanza, B. Haefner, and A. Kraemer, \"Optimization of selective assembly and adaptive manufacturing by means of cyber-physical system based matching,\" CIRP Ann., vol. 64, no. 1, pp. 399-402, 2015, doi: 10.1016/j.cirp.2015.04.123. [40] S. H. Khajavi, J. Partanen, and J. Holmström, \"Additive manufacturing in the spare parts supply chain,\" Comput. Ind., vol. 65, no. 1, pp. 50-63, Jan. 2014, doi: 10.1016/j.compind.2013.07.008. [41] J. Pearl, Causality: Models, Reasoning, and Inference, 2nd ed. New York, NY, USA: Cambridge University Press, 2009. [42] E. Bareinboim and J. Pearl, \"Causal inference and the data-fusion problem,\" Proc. Natl. Acad. Sci. U.S.A., vol. 113, no. 27, pp. 7345-7352, Jul. 2016, doi: 10.1073/pnas.1510507113. [43] M. F. Ashby and D. R. H. Jones, Engineering Materials 1: An Introduction to Properties, Applications and Design, 5th ed. Oxford, UK: Butterworth-Heinemann, 2019. [44] J. A. Schey, Introduction to Manufacturing Processes, 3rd ed. New York, NY, USA: McGraw-Hill, 2000. [45] R. Kacker, \"Economic considerations in designing experiments for evaluating the impact of reducing measurement uncertainty on product quality,\" J. Res. Natl. Inst. Stand. Technol., vol. 122, no. 33, pp. 1-14, Jun. 2017, doi: 10.6028/jres.122.033. [46] R. S. Kenett and S. Zacks, Modern Industrial Statistics: With Applications in R, MINITAB, and JMP, 2nd ed. Chichester, West Sussex, UK: John Wiley & Sons, 2014. [47] H. R. Varian, Intermediate Microeconomics: A Modern Approach, 9th ed. New York, NY, USA: W. W. Norton & Company, 2014. [48] J. Pearl, \"The seven tools of causal inference, with reflections on machine learning,\" Commun. ACM, vol. 62, no. 3, pp. 54-60, Feb. 2019, doi: 10.1145/3241036. [49] J. Y. Jung, G. Blau, J. F. Pekny, G. V. Reklaitis, and D. Eversdyk, \"A simulation based optimization approach to supply chain management under demand uncertainty,\" Comput. Chem. Eng., vol. 28, no. 10, pp. 2087-2106, Sep. 2004, doi: 10.1016/j.compchemeng.2004.06.006. [50] J. J. Michalek, F. M. Feinberg, and P. Y. Papalambros, \"Linking marketing and engineering product design decisions via analytical target cascading,\" J. Prod. Innov. Manag., vol. 22, no. 1, pp. 42-62, Jan. 2005, doi: 10.1111/j.0737-6782.2005.00102.x. [51] F. Eberhardt and R. Scheines, \"Interventions and causal inference,\" Philos. Sci., vol. 74, no. 5, pp. 981-995, Dec. 2007, doi: 10.1086/525638. [52] J. Pearl, \"Theoretical impediments to machine learning with seven sparks from the causal revolution,\" arXiv preprint arXiv:1801.04016, 2018. [53] T. W. Simpson, Z. Siddique, and J. Jiao, Eds., Product Platform and Product Family Design: Methods and Applications. New York, NY, USA: Springer, 2006. [54] Y. Zhang et al., \"Leak detection and location for gas pipelines using acoustic emission sensors,\" IEEE Access, vol. 7, pp. 101140-101149, 2019, doi: 10.1109/ACCESS.2019.2930953. [55] H. A. Kishawy and H. A. Gabbar, \"Review of pipeline integrity management practices,\" Int. J. Press. Vessels Pip., vol. 87, no. 7, pp. 373-380, Jul. 2010, doi: 10.1016/j.ijpvp.2010.04.003. [56] M. Tur, E. Geerlings, and W. B. Jeong, \"A review of leak detection and localization methods for liquid pipelines,\" J. Hydroinf., vol. 21, no. 6, pp. 1031-1052, Nov. 2019, doi: 10.2166/hydro.2019.012. [57] S. El-Zahab, T. Zayed, and M. Abdelkader, \"Leak detection in gas pipelines using artificial neural networks,\" J. Pipeline Syst. Eng. Pract., vol. 8, no. 1, p. 04016017, Feb. 2017, doi: 10.1061/(ASCE)PS.1949-1204.0000247. [58] S. Datta and S. Sarkar, \"A review on different pipeline fault detection methods,\" J. Loss Prev. Process Ind., vol. 41, pp. 97-106, May 2016, doi: 10.1016/j.jlp.2016.03.008. [59] J. Pearl, Causality: Models, Reasoning, and Inference, 2nd ed. New York, NY, USA: Cambridge University Press, 2009. [60] E. Bareinboim and J. Pearl, \"Causal inference and the data-fusion problem,\" Proc. Natl. Acad. Sci. U.S.A., vol. 113, no. 27, pp. 7345-7352, Jul. 2016, doi: 10.1073/pnas.1510507113. [61] C. Verde, L. Torres, and R. Obregón, \"A multi-leak detection and isolation method in pipelines based on a nonlinear extended Kalman filter,\" J. Process Control, vol. 52, pp. 55-68, Apr. 2017, doi: 10.1016/j.jprocont.2017.01.006. [62] J. Pearl, \"The seven tools of causal inference, with reflections on machine learning,\" Commun. ACM, vol. 62, no. 3, pp. 54-60, Feb. 2019, doi: 10.1145/3241036. [63] Y. Xu, J. Bao, J. Du, and X. Hu, \"Leak detection and location based on the flow similarity analysis of gas pipelines,\" J. Pet. Sci. Eng., vol. 176, pp. 1128-1140, May 2019, doi: 10.1016/j.petrol.2019.02.010. [64] M. S. Javadiha, S. Fathi, and H. Ghanbari, \"A new method for leak detection and location in gas pipelines based on pressure waves reflected from the leak point,\" J. Nat. Gas Sci. Eng., vol. 58, pp. 31-41, Oct. 2018, doi: 10.1016/j.jngse.2018.07.025. [65] X. Li, G. Chen, and H. Zhang, \"Leak detection and location of pipelines based on LMD and least squares twin support vector machines,\" IEEE Access, vol. 5, pp. 8659-8668, 2017, doi: 10.1109/ACCESS.2017.2693447. [66] X. Qin, X. Xiang, Y. Feng, and L. Wang, \"Improved pipeline leakage detection and location method based on leakage acoustic waves,\" IEEE Access, vol. 8, pp. 2111-2121, 2020, doi: 10.1109/ACCESS.2019.2962410. [67] Y. Li, W. Zhang, Z. Xiong, and G. He, \"Pipeline leak detection and location based on LSTM recurrent neural network,\" Sensors, vol. 20, no. 5, p. 1251, Feb. 2020, doi: 10.3390/s20051251 [68] J. Y. Halpern, Actual Causality. Cambridge, MA, USA: MIT Press, 2016. [69] T. Rasheed, O. San, and T. Kvamsdal, \"Digital twin: Values, challenges and enablers from a modeling perspective,\" IEEE Access, vol. 8, pp. 21980-22012, 2020, doi: 10.1109/ACCESS.2020.2970143. [70] S. A. Abdulrahman and A. J. Sultan, \"Artificial intelligence techniques in pipeline leak detection and condition assessment: A review,\" IEEE Access, vol. 9, pp. 62510-62542, 2021, doi: 10.1109/ACCESS.2021.3074255. [71] M. Mohri, A. Rostamizadeh, and A. Talwalkar, Foundations of Machine Learning, 2nd ed. Cambridge, MA, USA: MIT Press, 2018. [72] V. M. Bhaskaran and P. M. Narakasali, \"A review of inventory allocation strategies in supply chain management,\" Int. J. Adv. Res. Comput. Sci., vol. 8, no. 3, pp. 1129-1135, Mar. 2017, doi: 10.26483/ijarcs.v8i3.3135. [73] M. Gümü?, E. M. Jewkes, and J. H. Bookbinder, \"Impact of consignment inventory and vendor-managed inventory for a two-party supply chain,\" Int. J. Prod. Econ., vol. 113, no. 2, pp. 502-517, Jun. 2008, doi: 10.1016/j.ijpe.2007.10.019. [74] T. Choi and S. Sethi, \"Innovative quick response programs: A review,\" Int. J. Prod. Econ., vol. 127, no. 1, pp. 1-12, Sep. 2010, doi: 10.1016/j.ijpe.2010.05.010. [75] T. H. Truong and F. Azadivar, \"Optimal design methodologies for configuration of supply chains,\" Int. J. Prod. Res., vol. 43, no. 11, pp. 2217-2236, Jun. 2005, doi: 10.1080/00207540500050484. [76] I. Giannoccaro and P. Pontrandolfo, \"Supply chain coordination by revenue sharing contracts,\" Int. J. Prod. Econ., vol. 89, no. 2, pp. 131-139, May 2004, doi: 10.1016/S0925-5273(03)00047-1. [77] L. Zhang, \"Allocation of limited inventory to multiple customers with stochastic demands,\" Int. J. Logist. Syst. Manag., vol. 26, no. 3, pp. 315-330, Jan. 2017, doi: 10.1504/IJLSM.2017.081941. [78] K. Moinzadeh and S. Nahmias, \"A continuous review model for an inventory system with two supply modes,\" Manag. Sci., vol. 34, no. 6, pp. 761-773, Jun. 1988, doi: 10.1287/mnsc.34.6.761. [79] J. A. Muckstadt, \"Inventory control: An overview,\" in Handbook of Stochastic Models and Analysis of Manufacturing System Operations, J. G. Shanthikumar and D. D. Yao, Eds. New York, NY, USA: Springer, 2013, pp. 149-172. [80] K. Govindan, M. N. Popiuc, and A. Diabat, \"Overview of coordination contracts within forward and reverse supply chains,\" J. Clean. Prod., vol. 47, pp. 319-334, May 2013, doi: 10.1016/j.jclepro.2013.02.001. [81] R. Ganeshan, \"Managing supply chain inventories: A multiple retailer, one warehouse, multiple supplier model,\" Int. J. Prod. Econ., vol. 59, no. 1-3, pp. 341-354, Mar. 1999, doi: 10.1016/S0925-5273(98)00115-7. [82] D. J. Thomas and P. M. Griffin, \"Coordinated supply chain management,\" Eur. J. Oper. Res., vol. 94, no. 1, pp. 1-15, Oct. 1996, doi: 10.1016/0377-2217(96)00098-7. [83] J. Heydari, K. Govindan, and R. Sadeghi, \"Reverse supply chain coordination under stochastic remanufacturing capacity,\" Int. J. Prod. Econ., vol. 202, pp. 1-11, Aug. 2018, doi: 10.1016/j.ijpe.2018.04.029. [84] J. Pearl, Causality: Models, Reasoning, and Inference, 2nd ed. New York, NY, USA: Cambridge University Press, 2009. [85] E. Bareinboim and J. Pearl, \"Causal inference and the data-fusion problem,\" Proc. Natl. Acad. Sci. U.S.A., vol. 113, no. 27, pp. 7345-7352, Jul. 2016, doi: 10.1073/pnas.1510507113. [86] J. Pearl, \"The seven tools of causal inference, with reflections on machine learning,\" Commun. ACM, vol. 62, no. 3, pp. 54-60, Feb. 2019, doi: 10.1145/3241036. [87] J. G. Shanthikumar and D. D. Yao, \"Inventory models with stochastic lead times,\" in Analysis and Modeling of Manufacturing Systems, S. B. Gershwin, Y. Dallery, C. T. Papadopoulos, and J. M. Smith, Eds. Boston, MA, USA: Springer, 2002, pp. 157-186. [88] J. Hu and Y. Zhao, \"Inventory control for multi-product systems with fixed ordering cost and shortage penalties,\" IEEE Access, vol. 7, pp. 124202-124219, 2019, doi: 10.1109/ACCESS.2019.2938377. [89] D. D. Eisenstein, \"Recovering cyclic schedules using dynamic produce-up-to policies,\" Oper. Res., vol. 53, no. 4, pp. 675-688, Jul.-Aug. 2005, doi: 10.1287/opre.1040.0200. [90] Z. Luo, X. Chen, X. Chen, and X. Wang, \"Optimal pricing policies for differentiated brands under different supply chain power structures,\" Eur. J. Oper. Res., vol. 259, no. 2, pp. 437-451, Jun. 2017, doi: 10.1016/j.ejor.2016.10.048. [91] D. Simchi-Levi, P. Kaminsky, and E. Simchi-Levi, Designing and Managing the Supply Chain: Concepts, Strategies, and Case Studies, 3rd ed. New York, NY, USA: McGraw-Hill, 2008. [92] J. Pearl, \"Theoretical impediments to machine learning with seven sparks from the causal revolution,\" arXiv preprint arXiv:1801.04016, 2018. [93] D. Geiger, T. Verma, and J. Pearl, \"Identifying independence in Bayesian networks,\" Networks, vol. 20, no. 5, pp. 507-534, Aug. 1990, doi: 10.1002/net.3230200504. [94] J. Pearl and D. Mackenzie, The Book of Why: The New Science of Cause and Effect. New York, NY, USA: Basic Books, 2018. [95] E. Bareinboim and J. Pearl, \"A general algorithm for deciding transportability of experimental results,\" J. Causal Inference, vol. 1, no. 1, pp. 107-134, Jan. 2013, doi: 10.1515/jci-2012-0004. [96] S. A. Lippman and K. F. McCardle, \"The competitive newsboy,\" Oper. Res., vol. 45, no. 1, pp. 54-65, Jan.-Feb. 1997, doi: 10.1287/opre.45.1.54. [97] X. Xu, \"Optimal policies for stochastic inventory systems with a minimum order quantity and linear costs,\" IEEE Trans. Autom. Control, vol. 64, no. 1, pp. 349-357, Jan. 2019, doi: 10.1109/TAC.2018.2828463. [98] R. Bhatnagar and A. S. Sohal, \"Supply chain competitiveness: Measuring the impact of location factors, uncertainty and manufacturing practices,\" Technovation, vol. 25, no. 5, pp. 443-456, May 2005, doi: 10.1016/j.technovation.2003.09.012. [99] M. Mohri, A. Rostamizadeh, and A. Talwalkar, Foundations of Machine Learning, 2nd ed. Cambridge, MA, USA: MIT Press, 2018. [100] M. Colledani, T. Tolio, A. Fischer, B. Iung, G. Lanza, R. Schmitt, and J. Váncza, \"Design and management of manufacturing systems for production quality,\" CIRP Ann., vol. 63, no. 2, pp. 773-796, 2014, doi: 10.1016/j.cirp.2014.05.002. [101] S. A. Mansouri, E. Aktas, and U. Besikci, \"Green scheduling of a two-machine flowshop: Trade-off between makespan and energy consumption,\" Eur. J. Oper. Res., vol. 248, no. 3, pp. 772-788, Feb. 2016, doi: 10.1016/j.ejor.2015.08.064. [102] G. A. Zsidisin and L. M. Ellram, \"An agency theory investigation of supply risk management,\" J. Supply Chain Manag., vol. 39, no. 3, pp. 15-27, Aug. 2003, doi: 10.1111/j.1745-493X.2003.tb00156.x. [103] C. Öztürk, S. Sar?o?lu, and B. Baral, \"A review of contracts in supply chains,\" in Intelligent and Fuzzy Techniques: Smart and Innovative Solutions, C. Kahraman, S. Cevik Onar, B. Oztaysi, and I. U. Sari, Eds. Cham, Switzerland: Springer, 2021, pp. 1085-1092. [104] W. J. Hopp and M. L. Spearman, Factory Physics, 3rd ed. Long Grove, IL, USA: Waveland Press, 2011. [105] Y. Hou, W. Wang, and J. H. Lambert, \"Resilience analytics for identifying maintenance priorities and optimizing scheduling of manufacturing systems,\" IEEE Trans. Eng. Manag., vol. 68, no. 5, pp. 1481-1495, Oct. 2021, doi: 10.1109/TEM.2020.2978366. [106] P. R. Kleindorfer and G. H. Saad, \"Managing disruption risks in supply chains,\" Prod. Oper. Manag., vol. 14, no. 1, pp. 53-68, Mar. 2005, doi: 10.1111/j.1937-5956.2005.tb00009.x. [107] T. J. Kull and S. Talluri, \"A supply risk reduction model using integrated multicriteria decision making,\" IEEE Trans. Eng. Manag., vol. 55, no. 3, pp. 409-419, Aug. 2008, doi: 10.1109/TEM.2008.922627. [108] J. Pearl, Causality: Models, Reasoning, and Inference, 2nd ed. New York, NY, USA: Cambridge University Press, 2009. [109] E. Bareinboim and J. Pearl, \"Causal inference and the data-fusion problem,\" Proc. Natl. Acad. Sci. U.S.A., vol. 113, no. 27, pp. 7345-7352, Jul. 2016, doi: 10.1073/pnas.1510507113. [110] J. Pearl and D. Mackenzie, The Book of Why: The New Science of Cause and Effect. New York, NY, USA: Basic Books, 2018. [111] J. Y. Jung, G. Blau, J. F. Pekny, G. V. Reklaitis, and D. Eversdyk, \"A simulation based optimization approach to supply chain management under demand uncertainty,\" Comput. Chem. Eng., vol. 28, no. 10, pp. 2087-2106, Sep. 2004, doi: 10.1016/j.compchemeng.2004.06.006. [112] S. Chopra and M. S. Sodhi, \"Managing risk to avoid supply-chain breakdown,\" MIT Sloan Manag. Rev., vol. 46, no. 1, pp. 53-61, Fall 2004. [113] D. Ivanov, A. Dolgui, and B. Sokolov, \"The impact of digital technology and Industry 4.0 on the ripple effect and supply chain risk analytics,\" Int. J. Prod. Res., vol. 57, no. 3, pp. 829-846, Feb. 2019, doi: 10.1080/00207543.2018.1488086. [114] S. Axsäter, Inventory Control, 3rd ed. Cham, Switzerland: Springer, 2015. [115] M. Kaur and M. Kumar, \"Forecasting the impact of information sharing on retail supply chain management,\" IEEE Trans. Eng. Manag., vol. 69, no. 3, pp. 797-806, Jun. 2022, doi: 10.1109/TEM.2020.3005294. [116] M. Gümü?, E. M. Jewkes, and J. H. Bookbinder, \"Impact of consignment inventory and vendor-managed inventory for a two-party supply chain,\" Int. J. Prod. Econ., vol. 113, no. 2, pp. 502-517, Jun. 2008, doi: 10.1016/j.ijpe.2007.10.019. [117] K. Nikolopoulos, S. Punia, A. Schäfers, C. Tsinopoulos, and C. Vasilakis, \"Forecasting and planning during a pandemic: COVID-19 growth rates, supply chain disruptions, and governmental decisions,\" Eur. J. Oper. Res., vol. 290, no. 1, pp. 99-115, Apr. 2021, doi: 10.1016/j.ejor.2020.08.001. [118] S. Hosseini, N. Morshedlou, D. Ivanov, M. D. Sarder, K. Barker, and A. Al Khaled, \"Resilient supplier selection and optimal order allocation under disruption risks,\" Int. J. Prod. Econ., vol. 213, pp. 124-137, Jul. 2019, doi: 10.1016/j.ijpe.2019.03.018. [119] B. Çelik Nitaj, M. C. van der Heijden, and M. J. Land, \"Designing contracts for inventory management under uncertain quality and quantity,\" Eur. J. Oper. Res., vol. 300, no. 2, pp. 544-560, Apr. 2022, doi: 10.1016/j.ejor.2021.09.027. [120] D. Bandaly, A. Satir, Y. Kahyaoglu, and L. Shanker, \"Supply chain risk management - I: Conceptualization, framework and planning process,\" Risk Manag., vol. 14, no. 4, pp. 249-271, Nov. 2012, doi: 10.1057/rm.2012.7.

Copyright

Copyright © 2024 Sai Anvesh Durvasula. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.