Single Machine Production System Research Articles

Post-prognostics demand management, production, spare parts and maintenance planning for a single-machine system using Reinforcement Learning

Production Planning and Control (PPC) is crucial for any manufacturer and comprises steps such as demand management, production, or source planning. Manufacturers achieve competitive advantage by sustaining continuous production, which can be realised through Condition-based Maintenance and Prognostics and Health Management. Hereby, the machine’s health can be predicted, and post-prognostics decision-making allows to optimise PPC to meet customer demands and minimise costs. Unfortunately, the complex dynamic, stochastic and intransparent nature of post-prognostics PPC makes it intractable to use ‘traditional’ static or deterministic optimisation techniques or approaches that require an exact mathematical model or objective function. To tackle this, a data-driven post-prognostics Reinforcement Learning model is developed to plan and control the sourcing of spare parts, production, and maintenance of a single-machine production system to maximise production revenue by meeting customer demands and minimising costs. In a case study, Proximal Policy Optimisation, which is well-known from OpenAI’s ChatGPT, is applied to a post-prognostics PPC decision-making problem. The Proximal Policy Optimisation is compared to other state-of-the-art learners, and the performance and robustness are evaluated. Analyses show that our model outperforms other learners, as well as reactive and scheduled preventive maintenance strategies and is robust to noise and cost changes.

Joint optimisation of the maintenance and buffer stock policies considering back orders

Inventory control and maintenance systems are two main processes in production systems that play a vital role in the efficiency of production resources. This study aims to propose a joint optimisation procedure for maintenance and inventory policies in a single-machine production system considering back orders. The failure rate is random and the minimum level of accessibility required for preventive maintenance is considered. To close the problem at hand to the real-world condition, demand is considered as uncertain and shortage is inevitable. In addition, the unsatisfied demand is considered as back order. The problem is described and formulated via a mathematical model and all the possible scenarios corresponding to the different possible cases of the problem are described. Then a numerical iterative procedure is introduced to solve the problem in a reasonable time. The proposed procedure does not explore all solution space, but it behaves like a numerical optimisation procedure and provides a near optimal solution. The performance of the proposed method is evaluated using a numerical instance based on a real case study. In addition, some sensitivity analyses are performed to identify the most important effective parameters.

Integrated maintenance scheduling inventory policy adjustment considering back order disruptions using joint optimization approach

PurposeThe purpose of this paper is to present an integrated strategy for inventory control and preventive maintenance planning for a single-machine production system with increasing failure rates.Design/methodology/approachThere are three scenarios for solving presented model. The strategy is such that the production component is placed under maintenance as soon as it reaches the m level or in the event of a malfunction earlier than m. Maintenance completion time is not predictable. As a result of periodic maintenance, a buffer stock h is held and the production component starts to produce from period A with the maximum throughput to satisfy demand and handle the shortage. A numerical algorithm to find the optimal policy is developed. The algorithm is implemented using MATLAB software.FindingsThe authors discovered that joint optimization mainly reduces production system costs. Cs is holding cost of a product unit during a unit of time. The authors consider two values for Cs, consist of, Cs = 1 and Cs = 2. By comparing the two cases, it is concluded that by reducing the cost from Cs = 2 to Cs = 1, the optimal scenario does not differ. The amount of decision variables decreases.Originality/valueThis paper is the provision of a model in which the shortage of back order type is considered, which greatly increases the complexity of the problem compared to similar issues. The methods for solving such problems are provided by the numerical algorithm, and the use of buffers as a way to compensate for the shortage in the event of a complete shutdown of the production line which is a very effective and efficient way to deal with customer loss.

ON TIME-TO-BUFFER OVERFLOW DISTRIBUTION IN A SINGLE-MACHINE DISCRETE-TIME SYSTEM WITH FINITE CAPACITY

A model of a single-machine production system with finite magazine capacity is investigated. The input flow of jobs is organized according to geometric distribution of interarrival times, while processing times are assumed to be generally distributed. The closed-form formula for the generating function of the time to the first buffer overflow distribution conditioned by the initial buffer state is found. The analytical approach based on the idea of embedded Markov chain, the formula of total probability and linear algebra is applied. The corresponding result for next buffer overflows is also given. Numerical examples are attached as well.

Note on “Multiproduct Single-Machine Production System with Stochastic Scrapped Production Rate, Partial Backordering and Service Level Constraint”

An economic production quantity (EPQ) inventory model for a multiproduct single-machine production system with stochastic scrapped production rate, partial backordering and service level constraint is proposed. This research work revisits this EPQ inventory model and identifies some technical inconsistencies in it. Additionally, some observations and corrections to the shortcomings are provided as well as the solution procedure is improved.

Crane scheduling in a single machine production system with multiple LIFO buffers

Crane scheduling in a single machine production system with multiple LIFO buffers

Integrated production quality and condition-based maintenance optimisation for a stochastically deteriorating manufacturing system

This paper investigates the problem of optimally integrating production quality and condition-based maintenance in a stochastically deteriorating single- product, single-machine production system. Inspections are periodically performed on the system to assess its actual degradation status. The system is considered to be in ‘fail mode’ whenever its degradation level exceeds a predetermined threshold. The proportion of non-conforming items, those that are produced during the time interval where the degradation is beyond the specification threshold, are replaced either via overtime production or spot market purchases. To optimise preventive maintenance costs and at the same time reduce production of non-conforming items, the degradation of the system must be optimally monitored so that preventive maintenance is carried out at appropriate time intervals. In this paper, an integrated optimisation model is developed to determine the optimal inspection cycle and the degradation threshold level, beyond which preventive maintenance should be carried out, while minimising the sum of inspection and maintenance costs, in addition to the production of non-conforming items and inventory costs. An expression for the total expected cost rate over an infinite time horizon is developed and solution method for the resulting model is discussed. Numerical experiments are provided to illustrate the proposed approach.

An Imperfect Multi-Item Single Machine production system with Shortage, Rework, and Scrapped Considering Inspection, Dissimilar Deficiency Levels, and Non-Zero Setup Times

In this paper, we consider a deficient production system with permissible shortages. The production system consists of a unique machine that manufactures a number of products that a part of them are imperfect in form of rework or scrap. These defective products are identified by 100% inspection during production, then, they are whether reworked or disposed of after normal production process. Like real-world production systems, there are diverse kinds of errors creating dissimilar breakdown severity and rework. Moreover, reworks have non-zero setup times that makes the problem closer to real-world instances where machines require some preparations before starting a new production cycle. Thus, we introduce an economic production quantity (EPQ) problem for an imperfect manufacturing system with non-zero setup times for rework items. The rework items are classified into several categories based on their type of failure and rework rate. The aim of this study is to obtain optimum production time and shortage in each period that minimizes inventory system costs including setup costs in reworks, holding costs, production costs, inspection costs, disposal costs, and shortage. Finally, a numerical example is proposed to assess efficiency and validation of the proposed algorithm.

Estimation of overall equipment effectiveness using simulation programme

A queuing system of the M/M/1/N type with cyclic failure-free and repair times and cyclic defective production is used as a model of a single machine production system. Production tasks arrive according to an exponential process and are served with normally disturbed processing time. Successive working (failure-free) times, repair times and defect-free production have exponential distributions. Defective products can be recirculated and back to the production line according to a Bernoulli distributions. The objective is to evaluate effectiveness of the production system using Overall Equipment Effectiveness before and after introducing improvement initiatives.

Economic lot sizing for unreliable production system with shortages

The purpose of the present study is to analyse the optimal lot size in an unreliable single-machine production system with shortages. The production system is subject to failure due to machine breakdown. Breakdown times are considered to be according to Weibull distribution. It is assumed that the shortages are allowed and backlogged. During each production, the set-up preventive (regular) maintenance is performed. The corrective (i.e., emergency) maintenance is carried out immediately after breakdown. For the illustration purpose, numerical results are provided for the special cases. To obtain the optimal cost per unit time, we also employ the artificial neuro-fuzzy inference system (ANFIS) approach which has the learning capability of neural network as well as advantages of rule-base fuzzy system. It is noted that the results obtained by neuro-fuzzy technique are at par with the results computed by the analytical techniques.

Multiproduct single-machine production system with stochastic scrapped production rate, partial backordering and service level constraint

In this paper, a multiproduct single-machine production system under economic production quantity (EPQ) model is studied in which the existence of only one machine causes a limited production capacity for the common cycle length of all products, the production defective rates are random variables, shortages are allowed and take a combination of backorder and lost sale, and there is a service rate constraint for the company. The aim of this research is to determine the optimal production quantity, the allowable shortage level, and the period length of each product such that the expected total cost, including holding, shortage, production, setup and defective items costs, is minimized. The mathematical model of the problem is derived for which the objective function is proved to be convex. Then, a derivative approach is utilized to obtain the optimal solution. Finally, two numerical examples in each of which a sensitivity analysis is performed on the model parameters, are provided to illustrate the practical usage of the proposed methodology.

Sequence and due date assignment with uncertain processing times and mixed penalty functions

This research considers a single machine production system with stochastic processing times. Since complete distribution information of the processing time of each job is difficult or impossible to achieve, we introduce a heuristic procedure using only mean and variance information. An objective function combining three penalties: quadratic penalty on job earliness, quadratic penalty on job tardiness and linear penalty associated with long due date assignment is studied. Numerical tests indicate that our procedure performs better than the method only considering mean information. Furthermore, the performance of the proposed procedure is robust to job processing time distributions.

Lot sizing and scheduling with sequence-dependent setup costs and times and efficient rescheduling opportunities

This paper deals with lot sizing and scheduling for a single-stage, single-machine production system where setup costs and times are sequence dependent. A large-bucket mixed integer programming (MIP) model is formulated which considers only efficient sequences. A tailor-made enumeration method of the branch-and-bound type solves problem instances optimally and efficiently. The size of solvable cases ranges from 3 items and 15 periods to 10 items and 3 periods. Furthermore, it will become clear that rescheduling can neatly be done.

A comparison of Reorder Point and Kanban policies for a single machine production system

This paper examines two different policies for managing the production of a set of different parts on a single machine. The demands for the parts are uncertain, but have no lumpy order size. The Reorder Point and Kanban policies are tested in a range of environments characterized by different levels of demand variability, setup time requirement and machine utilization. The results show that Kanban requires less inventory than Reorder Point to achieve the same level of customer service. The dominance of Kanban over Reorder Point, however, diminishes as the demand variability, setup time or machine utilization is reduced. The findings also suggest that Kanban is a simpler policy for achieving different levels of customer service.

A Heuristic Scheduling Policy for Multi-Item, Multi-Machine Production Systems with Time-Varying, Stochastic Demands

An effective scheduling policy known as the Dynamic Cycle Lengths Heuristic was introduced by Leachman and Gascon in 1988 for the multi-item, single-machine production system facing stochastic, time-varying demands. In this article we develop a heuristic scheduling policy for the multi-machine extension of the same problem. We integrate the concepts of the Dynamic Cycle Lengths Heuristic with a nonlinear integer optimization model to obtain an overall scheduling policy that allocates items to machines and schedules production quantities during the next time period. We report promising performance in limited simulation tests of the policy.

Types of flexibility in a single-machine production system

Flexibility is one of the most important aspects of production systems. In this paper, different approaches to this subject are summarized and it is argued that various aspects of flexibility can be unified in a hierarchical structure. A simple, one-machine system is used to obtain the 'volume flexibility' from the 'machine flexibility' in the context of the hierarchical structure of flexibility types.

On-Line Scheduling in the Production Environment: Part I. An “Unsuccessful” Case History

A case history is presented from a 7,000 employee automotive transmission plant. It concerns a large transfer line department which could be viewed as a single machine production system with sequence-dependent setups. The objective is to identify those elements that are necessary for success in implementing on-line systems with operating personnel in a production environment. This case is the first of a two-part series; Part II will detail a successful implementation.