Off-line inspections under inspection errors

Coordinated procurement/inspection and production model under inspection errors

Optimal radio access technology selection on heterogeneous networks

Optimal production and product inspection policies for a deteriorating production system are considered. In theoretical investigations and practical applications using simplified models, inspection results may be assumed to be perfect. In this study, two kinds of inspection errors are considered. We first show that an inspection policy that does not inspect the first s produced items but inspects only those items from the ( s +1)th till the end of the production run is optimal. Two commonly used inspection strategies, no inspection and full inspection, are discussed. Next, we show how the optimal products inspection policy are incorporated into the determination of optimal lot size. Under the optimal inspection policy, the optimal production lot size is shown to be unique and bounded above by the traditional economic manufacturing quantity. Based on these findings, an efficient solution procedure is provided for the simultaneous determination of the optimal production lot size and its corresponding product inspection policy. Numerical examples are given to illustrate the model. Sensitivity analyses are performed to examine the effect of inspection errors on the solution.

Optimal and Near Optimal Control of a Two-Part-Type Stochastic Manufacturing System with Dynamic Setups

Optimum stochastic inspection policies for a system with safety device

Certain continuous production processes and early warning systems must be inspected in order to detect failures that would otherwise not be apparent. An optimal inspection policy is a sequence of inspection times which minimizes the average cost per inspection cycle relative to some cost model. Under both an existing cost model, and a new cost model proposed in this paper, the computation of such policies is difficult. Three sub-optimal inspection policies are proposed, each, like the optimal policy, having decreasing times between successive inspections when the time to system failure has increasing hazard rate. Numerical comparisons are given for both cost models in the case of Weibull failure times.

This paper is concerned with an easily computable inspection policy for the detection of the failure of a system. A one-parameter policy is suggested and it is shown that this policy has the property of decreasing (increasing) intervals between successive inspection times if the system has an increasing (decreasing) failure rate. Cost comparisons, using linear cost functions, with a computationally difficult optimal policy which has a minimium expected cost show that the proposed policy compares quite well with the optimal policy.

Optimal inspection policies: A review and comparison

A “phantom stockout" is a retail stockout phenomenon caused by unobserved inventory shrinkage. Unlike conventional stockouts that can be corrected by inventory replenishment, a phantom stockout persists and requires human inspection. In this study, we formulate such a problem as an infinite‐horizon Bayesian dynamic program with joint inventory inspection and replenishment decisions. This problem is challenging to solve due to non‐convexity and high dimensionality. However, we find that under the Bernoulli shrinkage process, the optimal inventory inspection policy has a simple threshold structure that depends on the number of consecutive zero‐sales periods since the last inspection, while the optimal inventory replenishment policy is the same as the optimal policy without inventory shrinkage. Our numerical studies further demonstrate that this simple and intuitive policy can be an effective heuristic for more general shrinkage processes.

This study considers a periodic‐review joint pricing and inventory control problem for a single product, where production incurs a fixed cost plus a convex or concave variable cost. Our objective is to maximize the expected discounted profit over the entire planning horizon. We fully characterize the optimal policy for the single‐period problem. As the optimal policy for the multi‐period problem is too complicated to be implemented in practice, we develop well‐structured heuristic policies, and establish worst‐case performance bounds on the profit gap between the heuristic policies and the optimal policies. Numerical studies show that our heuristic policies perform extremely well. To further reveal the structural properties of the optimal policies, we also introduce two new concepts named κ‐convexity and sym‐ κ‐convexity, provide the associated preservation results, and then characterize the optimal policies.

It is well-known in the standard age replacement policy that a finite preventive replacement time does not exist when the failure time is exponential and the optimal preventive replacement time is nonrandom. It is shown that when the failure time is exponential, a finite time exists by introducing the shortage and excess costs. In addition, the random age replacement is proposed and similar discussions are made. Furthermore, the periodic and random inspection policies are taken up, and their optimal policies are shown to correspond theoretically to those of the age replacement ones. It is shown finally that when the random inspection cost is the half of the periodic one, two expected costs are almost the same.

Multi-product Dynamic Pricing with Limited Inventories under Cascade Click Model

Maximizing the throughput (or revenue generation) rate has become one of die most important criteria in die design and management of Flexible Manufacturing Systems (FMSs). This paper develops dynamic part-allocation policies for FMSs having finite storage capacity at each work station. Maximizing the throughput rate means that the resulting queueing network model has a state-dependent arrival process; therefore, product-form solutions do not hold. Consequently, several alternative modeling approaches are described and formulated for deriving die optimal part-routing policies. One of these optimal policies is based on a new initiated-suspension part-routing strategy. This strategy results in a reduced load on the material handling system while increasing die expected throughput (or revenue generation) rates of the manufacturing work stations. We also propose several efficient closed-loop heuristic policies that exploit die response structure of die optimal policies. These heuristic policies are of...

Biosensors are tiny wireless medical devices which are attached or implanted into the body of a human being or animal to monitor and control biological processes. They are distinguished from conventional sensors by their biologically derived sensing elements. Biosensors generate heat when they transmit their measurements and their temperature rises when recharged by electromagnetic energy. These phenomena translate to a temperature increase in the tissues surrounding the biosensors. If the temperature increase exceeds a certain threshold, the tissues might be damaged. In this paper, we discuss the problem of finding an optimal operating policy for a rechargable biosensor under a strict maximum temperature increase constraint. This problem can be formulated as a Markov decision process with an average reward criterion. The solution is an optimal policy that maximizes the average number of samples which can be generated by the biosensor while observing the constraint on the maximum safe temperature level. Due to the exponential nature of the problem, a heuristic policy is proposed. The performance of the policies is studied through simulation. A greedy policy is used as a baseline for comparison.

In this paper, we consider the problem of finding optimal energy management policies in the presence of energy harvesting sources to maximize network performance. We formulate this problem in the discounted cost Markov decision process framework and apply two reinforcement learning algorithms. Prior work obtains optimal policy in the case when the conversion function mapping energy to data transmitted is linear and provides heuristic policies in the case when the same is nonlinear. Our algorithms, however, provide optimal policies regardless of the form of the conversion function. Through simulations, our policies are seen to outperform those of in the nonlinear case.