arXiv subjects

Melvin Drent

Publications and source records attributed to Melvin Drent.

8 recordsLinked to original sources

arXiv2026

Risk or Replace: Efficient Asymptotics for Data-Driven Maintenance

Condition-based maintenance (CBM) is an approach that plans interventions for deteriorating systems according to their observed operational state. CBM reduces unplanned downtime and extends usable lifetime. We study a heterogeneous population of components that degrade over time according to a stochastic processes with non-negative and i.i.d. increments that are characterized by component-specific parameters that remain unobservable to the decision maker. We rely on degradation data to estimate these parameters and determine replacement actions at equidistant epochs. The goal is to minimize the long-run average cost, which incorporates fixed replacement costs, failure costs, and operating costs. This problem can be formulated as a high-dimensional partially observable Markov decision process (POMDP), which is generally intractable. We develop a tractable, data-driven CBM policy that estimates the optimal policy of a hypothetical Oracle that has full information of the underlying degradation parameters and call this policy the Estimated Oracle's Optimal Policy (EOP). We introduce a scaling regime where both the failure thresholds and cost parameters increase proportionally, reflecting practical settings in which component lifetimes and maintenance costs are large relative to the time between two consecutive CBM decision moments. We show that the regret of the EOP, defined as the difference between its long-run average cost and that of the Oracle, converges to zero in the scaling regime when the parameter estimator is consistent. Across extensive experiments using both real and simulated data, the EOP achieves very low regret and, whenever the optimal POMDP policy can be computed exactly, a negligible optimality gap.

math.OC↗

arXiv2025

Same-day or next-day? Transparent time-dependent shipment pricing for e-fulfillment

We develop a parsimonious model of an e-commerce fulfillment center that offers time-dependent shipment options and corresponding fees to utility-maximizing customers arriving according to a Poisson process. For any such policy, we provide an exact steady-state analysis using the underlying periodic Markov chain to characterize system performance. Because shipment fees shape both the volume and timing of same-day demand, direct optimization over the price domain is analytically intractable. To enable structural and computational insights, we introduce a transformation that maps each shipment-fee policy to its induced cumulative demand profile. This reformulation reveals that the optimal policy features a cutoff time and monotonically increasing fees, and it yields a supermodular profit function that can be optimized in polynomial time. We also propose a simple two-level time-dependent fee structure that is intuitive for customers and achieves near-optimal performance. Numerical experiments show that introducing a cutoff time substantially improves profits under static fees, and that using time-dependent fees produces further significant gains. Overall, transparent time-dependent shipment policies help firms align same-day demand with fulfillment capacity while maintaining transparency and fairness for customers.

math.OC↗

arXiv2025

Asymptotic Optimality of Projected Inventory Level Policies for Lost Sales Inventory Systems with Large Leadtime and Penalty Cost

We study the canonical periodic review lost sales inventory system with positive leadtime and independent and identically distributed (i.i.d.) demand under the average cost criterion. We demonstrate that the relative value function under the constant order policy satisfies the Wiener-Hopf equation. We employ ladder processes associated with a random walk featuring i.i.d. increments, to obtain an explicit solution for the relative value function. This solution can be expressed as a quadratic form and a term that grows sublinearly. Then we perform an approximate policy iteration step on the constant order policy and uniformly bound the gap relative to the otimal cost rate for large lead times. This leads to our main result that projected inventory level policies are asymptotically optimal as the leadtime grows when the cost of losing a sale is sufficiently large and demand has a finite second moment.

math.PR↗

arXiv2023

Condition-Based Production for Stochastically Deteriorating Systems: Optimal Policies and Learning

Production systems deteriorate stochastically due to usage and may eventually break down, resulting in high maintenance costs at scheduled maintenance moments. This deterioration behavior is affected by the system's production rate. While producing at a higher rate generates more revenue, the system may also deteriorate faster. Production should thus be controlled dynamically to trade-off deterioration and revenue accumulation in between maintenance moments. We study systems for which the relation between production and deterioration is known and the same for each system as well as systems for which this relation differs from system to system and needs to be learned on-the-fly. The decision problem is to find the optimal production policy given planned maintenance moments (operational) and the optimal interval length between such maintenance moments (tactical). For systems with a known production-deterioration relation, we cast the operational decision problem as a continuous-time Markov decision process and prove that the optimal policy has intuitive monotonic properties. We also present sufficient conditions for the optimality of bang-bang policies and we partially characterize the structure of the optimal interval length, thereby enabling efficient joint optimization of the operational and tactical decision problem. For systems that exhibit variability in their production-deterioration relations, we propose a Bayesian procedure to learn the unknown deterioration rate under any production policy. Our extensive numerical study indicates significant profit increases of our approaches compared to the state-of-the-art.

math.OC↗

arXiv2023

Optimal data pooling for shared learning in maintenance operations

We study optimal data pooling for shared learning in two common maintenance operations: condition-based maintenance and spare parts management. We consider a set of systems subject to Poisson input -- the degradation or demand process -- that are coupled through an a-priori unknown rate. Decision problems involving these systems are high-dimensional Markov decision processes (MDPs) and hence notoriously difficult to solve. We present a decomposition result that reduces such an MDP to two-dimensional MDPs, enabling structural analyses and computations. Leveraging this decomposition, we (i) demonstrate that pooling data can lead to significant cost reductions compared to not pooling, and (ii) show that the optimal policy for the condition-based maintenance problem is a control limit policy, while for the spare parts management problem, it is an order-up-to level policy, both dependent on the pooled data.

cs.LG↗

arXiv2023

Dedicated maintenance and repair shop control for spare parts networks

We study a repairable inventory system dedicated to a single component that is critical in operating a capital good. The system consists of a stock point containing spare components, and a dedicated repair shop responsible for repairing damaged components. Components are replaced using an age-replacement strategy, which sends components to the repair shop either preventively if it reaches the age-threshold, and correctively otherwise. Damaged components are replaced by new ones if there are spare components available, otherwise the capital good is inoperable. If there is free capacity in the repair shop, then the repair of the damaged component immediately starts, otherwise it is queued. The manager decides on the number of repairables in the system, the age-threshold, and the capacity of the repair shop. There is an inherent trade-off: A low (high) age-threshold reduces (increases) the probability of a corrective replacement but increases (decreases) the demand for repair capacity, and a high (low) number of repairables in the system leads to higher (lower) holding costs, but decreases (increases) the probability of downtime. We first show that the single capital good setting can be modelled as a closed queuing network with finite population, which we show is equivalent to a single queue with fixed capacity and state-dependent arrivals. For this queue, we derive closed-form expressions for the steady-state distribution. We subsequently use these results to approximate performance measures for the setting with multiple capital goods.

math.PR↗

arXiv2023

Efficient Emission Reduction Through Dynamic Supply Mode Selection

We study the inbound supply mode and inventory management decision making for a company that sells an assortment of products. Stochastic demand for each product arrives periodically and unmet demand is backlogged. Each product has two distinct supply modes that may be different suppliers or different transport modes from the same supplier. These supply modes differ in terms of their carbon emissions, speed, and costs. The company needs to decide when to ship how much using which supply mode such that total holding, backlog, and procurement costs are minimized while the emissions associated with different supply modes across the assortment remains below a certain target level. Since the optimal policy for this inventory system is highly complex, we assume that shipment decisions for each product are governed by a dual-index policy. This policy dynamically prescribes shipment quantities with both supply modes based on the on-hand inventory, the backlog, and the products that are still in-transit. We formulate this decision problem as a mixed integer linear program that we solve through Dantzig-wolfe decomposition. We benchmark our decision model against two state-of-the-art approaches in a large test-bed based on real-life carbon emissions data. Relative to our decision model, the first benchmark lacks the flexibility to dynamically ship products with two supply modes while the second benchmark makes supply mode decisions for each product individually rather than holistically for the entire assortment. Our computational experiment shows that our decision model can outperform the first and second benchmark by up to 15 and 40 percent, respectively, for realistic targets for carbon emission reduction.

math.OC↗

arXiv2022

Effective Dual-Sourcing Through Inventory Projection

We consider a single-echelon inventory system under periodic review with two suppliers facing stochastic demand, where excess demand is backlogged. The expedited supplier has a shorter lead time than the regular supplier but charges a higher unit price. We introduce the Projected Expedited Inventory Position (PEIP) policy, and we show that the relative difference between the long run average cost per period of this policy and the optimal policy converges to zero when both the shortage cost and the cost premium for expedited units become large, with their ratio held constant. A corollary of this result is that several existing heuristics are also asymptotically optimal in this non-trivial regime. We show through an extensive numerical investigation that the PEIP policy outperforms the current best performing heuristic policies in literature.

math.OC↗