Supply Chain Analytics最新文献

Electric vertical takeoff and landing (eVTOL) aircraft manufacturers await numerous pre-orders for eVTOLs and expect demand for such advanced air mobility (AAM) aircraft to rise dramatically soon. However, eVTOL manufacturers (EMs) cannot commence mass production of commercial eVTOLs due to a lack of supply chain planning for eVTOL manufacturing. The eVTOL supply chain differs from traditional ones due to stringent quality standards and limited suppliers for eVTOL parts, shortages in skilled labor and machinery, and contract renegotiations with major aerospace suppliers. The emerging AAM aircraft market introduces uncertainties in supplier pricing and capacities, eVTOL manufacturing costs, and eVTOL demand, further compounding the supply chain planning challenges for EMs. Despite this critical need, no study has been conducted to develop a comprehensive supply chain planning model for EMs. To address this research gap, we propose a stochastic optimization model for integrated supply chain planning of EMs while maximizing their operating profits under the abovementioned uncertainties. We conduct various numerical cases to analyze the impact of 1) endogenous eVTOL demand influenced by the quality of eVTOLs, 2) supply chain disruptions caused by geopolitical conflicts and resource scarcity, and 3) high-volume eVTOL demand similar to that experienced by automotive manufacturers, on EM supply chain planning. The results indicate that our proposed model is adaptable in all cases and outperforms established benchmark stochastic models. The findings suggest that EMs can commence mass eVTOL production with our model, enabling them to make optimal decisions and profits even under potential disruptions.

Precise demand forecasting and agile pricing strategies are crucial in modern business. This study aims to enhance these strategies by evaluating the efficacy of Holt-Winters Exponential Smoothing (HWES) and Autoregressive Integrated Moving Average (ARIMA) models. The study assesses their performance in predicting demand amid unpredictable factors and develops robust forecasting algorithms using real-world data. It evaluates HWES and ARIMA in capturing demand fluctuations, considering seasonality, market trends, and cyclic patterns. A comprehensive comparative analysis is conducted under stable and unstable economic conditions. The study also focuses on a dynamic pricing model for limited sale seasons, examining lost sales patterns over time. In the context of supply chain and inventory management, efficient demand forecasting and dynamic pricing are essential for optimizing inventory levels and minimizing costs. Supply chains must adapt quickly to demand fluctuations to avoid overstocking or stockouts, which lead to revenue losses and inefficiencies. The findings reveal that ARIMA consistently outperforms HWES in minimizing lost sales, demonstrating its efficacy in demand forecasting, mitigating stockouts, and reducing revenue losses, particularly in varying economic conditions. This research significantly contributes to current knowledge by developing tailored forecasting algorithms and a dynamic pricing model, enhancing supply chain resilience and performance in uncertain business environments.

Maintenance strategies are vital for industrial and manufacturing systems. This study considers a proactive maintenance strategy and emphasizes using analytics and data science. We propose an Explainable Artificial Intelligence (XAI) methodology for predictive maintenance. The proposed method utilizes a machine learning project cycle and Python libraries to interpret the results using the Local Interpretable Model-agnostic Explanations (LIME) method. We also introduce an early concept of spare parts management, presenting insights from predictive maintenance outcomes and providing explanations for decision-makers to enhance their understanding of the influential factors behind predictions. This study demonstrates that utilizing machine learning models in predictive maintenance is highly beneficial; however, the binary outcomes of these models can be misunderstood by decision-makers. Detailed explanations provided to decision-makers will directly impact maintenance decisions and improve spare part management.

Disasters are unforeseen occurrences requiring extensive transport deployment to support and relieve victims. Sometimes, this transportation is not feasible directly from some supply points to some destination points. Due to this tragedy, it is unclear precisely what is available at supply points, what is needed at destinations, how much transportation capacity there is, and what the routes are like. In this study, we investigate a two-stage multi-item fixed charge four-dimensional transportation problem using the concept of big data theory under the two-fold uncertainties. Here, the model’s parameters such as unit transportation costs, availabilities of items at the suppliers, fixed charges, capacities of conveyances, and demands of the items at the retailers are considered type-2 zigzag uncertain variables. Using big data theory and based on uncertain programming theory, two novel uncertain models are developed such as chance-constrained programming and expected value programming model. These two uncertain models transformed into the deterministic form via uncertainty inverse distribution theory. A critical value based reduction method with three categories (i.e., expected value, pessimistic value, and optimistic value) is applied to reduce the type-2 zigzag uncertain variable to the type-1 zigzag uncertain variable. The genetic algorithm and particle swarm optimization techniques have been proposed to find the optimal solution for the two deterministic models. The efficiency of our proposed approach is demonstrated with a real-life numerical example.

Home care services have a significant role in lowering healthcare expenditure. Supply chain management in home healthcare (HHC) ensures efficient delivery of medical supplies and equipment to patients' homes, improving overall quality of care and patient outcomes. This study proposes a routing and scheduling optimization model for HHC by prioritizing patients, developing an effective delivery strategy, and considering home care logistics and services. The model primarily concerns reducing logistics activities’ overall expenses while considering patients’ priorities. A bi-objective optimization model for a multi-period HHC problem is developed by prioritizing patients with urgent critical needs. The best-worst method (BWM) and technique for order of preference by similarity to ideal solution (TOPSIS) are used to prioritize patients using a linear programming metric (Lp-metric). The BWM and TOPSIS have been uniquely used in this study for routing and scheduling in HHC. Eventually, the applicability of the proposed method is demonstrated through a real-life case study with a series of numerical examples and sensitivity analysis. For instance, by analyzing privilege, we see patients are carefully matched with caregivers possessing advanced skills, leading to increased patient satisfaction. Based on assigned routes, caregivers prioritize patients with higher weight and emergency conditions at the start of each path, followed by patients with less urgent conditions. This ensures that patients with more severe conditions are serviced first.

Digital Twins (DTs) developments are still in the pilot stages of deployment in supply chain management (SCM), and their full integration with real-time synchronization and autonomous decision-making poses many challenges. This paper aims to identify these common challenges and provide a conceptual framework for establishing a Digital Twin (DT) system to improve supply chain management performance. The paper presents a systematic literature review of 129 research papers on DT applications for SCM improvement. The selected papers were reviewed and classified into three categories: manufacturing and production, supply chain, and logistics. The development of digital technologies such as the Internet of Things (IoT), Radio Frequency Identification (RFID) devices, cloud computing, cyber-physical systems (CPSs), cybersecurity (CS), and simulation modeling has increased the opportunities to explore the creation of supply chain DTs. However, there are limitations and various challenges due to the complexity of most systems. The results indicate that DT for SCM should include external links (i.e. suppliers, distributors) and internal links (i.e. procurement, production, logistics) to deal with any disruption through data-driven modeling with real-time synchronization. Based on the review findings, this study proposes a three-layered conceptual framework to improve supply chain management performance. The proposed framework provides future directions for DT research in SCM. It provides a holistic and integrated approach to DT implementation, the common DT technologies, and data analytics techniques for improved supply chain performance.

Despite abundant research on the application of artificial intelligence (AI) in reverse logistics, no comprehensive study with bibliometric and network analysis has been conducted. This study uses bibliometric analysis to derive the prominent research statistics in AI-centric reverse logistics, considering 2929 articles from the last three decades. The most impactful contributors and countries that employ AI in reverse logistics are identified using various bibliometric tools. Also, network analysis is performed to reveal the most influential articles and emerging trends and map the relationships via clustering. The results of keyword co-occurrence and co-citation analyses reveal that machine learning and deep learning techniques have been commonly used for addressing reverse logistics challenges with higher frequency in recent years. Furthermore, a systematic review is carried out, considering the influential articles from recent years. The review is conducted following the systematic literature review framework, and 79 articles are chosen to be studied thoroughly. Subsequently, the articles are divided based on various reverse logistics processes, and the most frequently used AI techniques are identified and categorized into five distinct groups. The comprehensive investigation of AI techniques reveals the use-case scenario of AI algorithms in the reverse logistics domain. This study concludes with implications and recommendations for prospects by addressing the shortcomings of the current studies and providing future researchers and practitioners with a robust roadmap to investigate reverse logistics in their research further.

Supplier selection has become increasingly complex regarding selection criteria caused by expanded data collection processes and supplier numbers due to globalisation effects. This complexity has led to the consideration of Artificial Intelligence (AI) techniques to facilitate and enhance supplier selection. However, the AI techniques most often applied are unfamiliar to stakeholders and have limited explainability, posing a significant barrier to adopting intelligent approaches in supply chains. To address this issue, we propose a hybrid supplier selection framework that combines interpretable data-driven AI techniques with multi-criteria decision-making (MCDM) approaches: the former aims to reduce the complexity of the supplier selection problem, while the latter ensures familiarity to supply chain stakeholders by retaining MCDM at the heart of the supplier selection process. The framework is validated through two real-world case studies supporting supplier selection decisions in oil, gas, and aerospace manufacturing companies. Preliminary results from our case studies suggest that the framework can achieve comparable performance to approaches utilising only machine learning while offering the added benefits of end-to-end explainability and increased familiarity.