This study proposes a Reinforcement Learning-based replenishment strategy aimed at improving the responsiveness of a real-world case study from the semiconductor industry, operating in a complex environment characterized by perishable products, stochastic lead time, and minimum order quantity. The agent explicitly determines the optimal order quantity while managing product perishability. To validate the effectiveness of the proposed replenishment policy, it has been compared with the Smoothing Order-Up-To (SOUT) policy. Such comparison analysis is based on two performance measures, such as the fill rate and the amount of expired products. For the sake of fairness, the SOUT replenishment parameters were calibrated by means of a Response Surface Methodology analysis. An extensive experimental campaign was carried out and numerical results show the outperformance of the Reinforcement Learning over the SOUT by mitigating the effect of variability in the lead time, which in turn implies inventory cost reduction and fill-rate increase.

Customs digitalisation single-window systems, pre-arrival processing, and automated risk management is widely promoted as a means of compressing the variability of border-clearance time. Drawing on classical inventory theory, we ask how that variance compression affects a firm’s optimal safety stock. Using a discrete-event simulation of a continuous-review (Q, r) replenishment system with skewed (lognormal) clearance times, and isolating the variance channel through a mean-preserving compression, we find that the inventory benefit of digitalisation is strongly regime-dependent. Above a cycle-service-level threshold (≈0.68 in our base case), variance compression reduces required safety stock substantially by roughly two-thirds at very high service levels. Below the threshold across the 50–70% band where most firms operate the benefit vanishes and reverses: variance compression raises the required buffer. The widely assumed inventory dividend of customs digitalisation thus accrues mainly to high-service-level firms. This regime-heterogeneity also offers a mechanism reconciling the operations prediction with the null clearance-uncertainty effect reported in aggregate trade data. The result speaks directly to the conference theme of moving “from data to policy”: the operational returns to a border-digitalisation policy depend on the service-level profile of the importing firm population.

This study develops a multi-item multi-supplier probabilistic inventory policy for an SME household-goods retailer by considering partial backorders and storage capacity constraints. The case covers 60 fast-moving SKUs supplied by eight fixed suppliers. Two policies are evaluated: a Supplier-Based Policy (SBP) with different ordering intervals across suppliers and an All-Supplier Policy (ASP) with one common interval per month. For historically stockout SKUs, the observed unmet-demand gap of 966 units is reduced to a modeled cumulative expected shortage of six units over seven months. Across all SKUs, both policies maintain monthly demand fulfillment above 95%. Replenishment quantities are adjusted to SKU-level demand and remaining inventory position, reducing unnecessary orders and excess-inventory risk. SBP yields the lowest calculated cost, while ASP offers a simpler schedule. Sensitivity analysis identifies storage capacity as the most influential parameter on the selected interval and total inventory cost.

Manufacturing is one of the biggest industries and supply chain is one of the key factors in staying competitive, however, it could be complex due to several externalities. The current study investigated operational struggles to shape logistics by balancing the complexity of risk management and how those decisions could affect the supplier relationship and how to mitigate problems due to constant market changes. A total of 200 respondents from the supply chain industry were surveyed using Structural Equation Modeling (SEM) to investigate the 10 latent variables. The results showed that technical and operational barriers make it challenging to stay sustainable, while economic difficulties and poor communication could result in management problems. The results corroborate the configurational and dynamic capabilities theories indicating that success depends on the organization’s internal capabilities on how they can easily address the market challenges.