Optimal Policy for a Stochastic Scheduling Problem with Applications to Surgical Scheduling

We consider the stochastic, single‐machine earliness/tardiness problem (SET), with the sequence of processing of the jobs and their due‐dates as decisions and the objective of minimizing the sum of the expected earliness and tardiness costs over all the jobs. In a recent paper, Baker ( 2014 ) shows the optimality of the Shortest‐Variance‐First (SVF) rule under the following two assumptions: (a) The processing duration of each job follows a normal distribution. (b) The earliness and tardiness cost parameters are the same for all the jobs. In this study, we consider problem SET under assumption (b). We generalize Baker's result by establishing the optimality of the SVF rule for more general distributions of the processing durations and a more general objective function. Specifically, we show that the SVF rule is optimal under the assumption of dilation ordering of the processing durations. Since convex ordering implies dilation ordering (under finite means), the SVF sequence is also optimal under convex ordering of the processing durations. We also study the effect of variability of the processing durations of the jobs on the optimal cost. An application of problem SET in surgical scheduling is discussed.     

MINIMIZING ABSOLUTE AND SQUARED DEVIATION OF COMPLETION TIMES FROM DUE DATES

This is a study of a single-machine scheduling problem with the objective of minimizing the sum of a function of earliness and tardiness called the earliness and tardiness (ET) problem. I will show that if priority weights of jobs are proportional to their processing times, and if earliness and tardiness cost functions are linear, the problem will be equivalent to the total weighted tardiness problem. This proves that the et problem is np -hard. In addition, I present a heuristic algorithm with worst case bound for the et problem based on the equivalence relation between the two. When earliness and tardiness cost functions are quadratic, I consider the problem for a common due date for all jobs and for different job due dates. In general, the et problem with quadratic earliness and tardiness cost functions and all job weights equal to one is np -hard. I show that in many cases, when weights of jobs are proportional to their processing times, the problem can be solved efficiently. In the published results on the et problem with quadratic earliness and tardiness cost functions other researchers have assumed a zero starting time for the schedule. I discuss the advantages of a nonzero starting time for the schedule.     

COMMON DUE-WINDOW SCHEDULING

In this paper, we solve common due-window scheduling problems within the just-in-time window concept, i.e., scheduling problems including both earliness and tardiness penalties. We assume that jobs share the same due window and incur no penalty as long as they are completed within the due window. We further assume that the earliness and tardiness penalty factors are constant and that the size of the window is a given parameter. For cases where the location of the due window is a decision variable, we provide a polynomial algorithm with complexity O(n * log (n)) to solve the problem. For cases where the location of the due window is a given parameter, we use dynamic programming with pseudopolynomial complexity to solve the problem.     

Approximation algorithm for the parallel-machine scheduling problem with release dates and submodular rejection penalties

In this paper, we consider the parallel-machine scheduling problem with release dates and submodular rejection penalties. In this problem, we are given m identical parallel machines and n jobs. Each job has a processing time and a release date. A job is either rejected, in which case a rejection penalty has to be paid, or accepted and processed on one of the m identical parallel machines. The objective is to minimize the sum of the makespan of the accepted jobs and the rejection penalty of the rejected jobs which is determined by a submodular function. Our main work is to design a 2-approximation algorithm based on the primal-dual framework.

     

TECHNICAL NOTE: A SIMPLE MODEL FOR OPTIMIZING THE SINGLE MACHINE EARLY/TARDY PROBLEM WITH SEQUENCE-DEPENDENT SETUPS

A simple mixed integer programming model for the N job/single machine scheduling problem with possibly sequence-dependent setup times, differing earliness/tardiness cost penalties, and variable due dates is proposed and evaluated for computational efficiency. Results indicated that the computational effort required to reach optimality rose with the number of jobs to be scheduled and with decreased variance in due dates. Though computational effort was significant for the largest problems solved, the model remained viable for optimizing research scale problems.     

A SINGLE-MACHINE SCHEDULING MODEL WITH FIXED-INTERVAL DELIVERIES

We consider an environment where a production facility modeled as a single machine needs to assign delivery dates to several orders and find a feasible sequence. Tardy jobs are not allowed. The delivery dates are to be at prespecified fixed intervals. The objective is to minimize the due date penalty and the cost of earliness. We provide a dynamic programming-based solution procedure that runs in polynomial time. We develop several dominance results that reduced the computational requirement by an order of magnitude in our computational study.     

Single-machine batch delivery scheduling with an assignable common due window

This paper addresses a batch delivery single-machine scheduling problem in which jobs have an assignable common due window. Each job will incur an early (tardy) penalty if it is early (tardy) with respect to the common due window under a given schedule. There is no capacity limit on each delivery batch, and the cost per batch delivery is fixed and independent of the number of jobs in the batch. The objective is to find the optimal size and location of the window, the optimal dispatch date for each job, as well as an optimal job sequence to minimize a cost function based on earliness, tardiness, holding time, window location, window size, and batch delivery. We show that the problem can be optimally solved in O(n⁸)$O(n^8)$ time by a dynamic programming algorithm under a reasonable assumption on the relationships among the cost parameters. A computational experiment is also conducted to evaluate the performance of the proposed algorithm. We also show that some special cases of the problem can be optimally solved by lower order algorithms.     

A FRAMEWORK FOR MACHINE SCHEDULING PROBLEMS WITH CONTROLLABLE PROCESSING TIMES

This is a study of single and parallel machine scheduling problems with controllable processing time for each job. The processing time for job j depends on the position of the job in the schedule and is a function of the number of resource units allocated to its processing. Processing time functions and processing cost functions are allowed to be nonlinear. The scheduling problems considered here have important applications in industry and include many of the existing scheduling models as special cases. For the single machine problem, the objective is minimization of total compression costs plus a scheduling measure. The scheduling measures include makespan, total flow time, total differences in completion times, total differences in waiting times, and total earliness and tardiness with a common due date for all jobs. Except when the total earliness and tardiness measure is involved, each case the problem is solved efficiently. Under an assumption typically satisfied in just-in-time systems, the problem with total earliness and tardiness measure is also solved efficiently. Finally, for a large class of processing time functions; parallel machine problems with total flow time and total earliness and tardiness measures are solved efficiently. In each case we reduce the problem to a transportation problem.     

A tardiness-augmented approximation scheme for rejection-allowed multiprocessor rescheduling

In the multiprocessor scheduling problem to minimize the total job completion time, an optimal schedule can be obtained by the shortest processing time rule and the completion time of each job in the schedule can be used as a guarantee for scheduling revenue. However, in practice, some jobs will not arrive at the beginning of the schedule but are delayed and their delayed arrival times are given to the decision-maker for possible rescheduling. The decision-maker can choose to reject some jobs in order to minimize the total operational cost that includes three cost components: the total rejection cost of the rejected jobs, the total completion time of the accepted jobs, and the penalty on the maximum tardiness for the accepted jobs, for which their completion times in the planned schedule are their virtual due dates. This novel rescheduling problem generalizes several classic NP-hard scheduling problems. We first design a pseudo-polynomial time dynamic programming exact algorithm and then, when the tardiness can be unbounded, we develop it into a fully polynomial time approximation scheme. The dynamic programming exact algorithm has a space complexity too high for truthful implementation; we propose an alternative to integrate the enumeration and the dynamic programming recurrences, followed by a depth-first-search walk in the reschedule space. We implemented the alternative exact algorithm in C and conducted numerical experiments to demonstrate its promising performance.

     

A heuristic algorithm for minimizing maintenance workforce level

We develop a heuristic algorithm for minimizing the workforce level required to accommodate all the maintenance jobs requested within a specific time interval. Each maintenance job has its own release and due dates as well as the required man-days, and must be scheduled in a noninterrupted time interval, i.e. without preemption. However, the duration of each job is not fixed, but to be determined within a specific range. We show that this problem can be seen as a variant of the two-dimensional bin-packing problem with some additional constraints. We develop a non-linear mixed integer programming model for the proposed problem, and employ some well-known bin-packing algorithms to develop an efficient heuristic algorithm. In order to evaluate the performance of the proposed heuristic, we present a computationally efficient scheme for getting a good lower bound for the actual minimum. The computational experiment shows that the proposed heuristic algorithm performs quite satisfactorily in practice.     

Minmax common flow-allowance problems with convex resource allocation and position-dependent workloads

We study minmax due-date based on common flow-allowance assignment and scheduling problems on a single machine, and extend known results in scheduling theory by considering convex resource allocation. The total cost function of a given job consists of its earliness, tardiness and flow-allowance cost components. Thus, the common flow-allowance and the actual jobs’ processing times are decision variables, implying that the due-dates and actual processing times can be controlled by allocating additional resource to the job operations. Consequently, our goal is to optimize a cost function by seeking the optimal job sequence, the optimal job-dependent due-dates along with the actual processing times. In all addressed problems we aim to minimize the maximal cost among all the jobs subject to a constraint on the resource consumption. We start by analyzing and solving the problem with position-independent workloads and then proceed to position-dependent workloads. Finally, the results are generalized to the method of common due-window. For all studied problems closed form solutions are provided, leading to polynomial time solutions.

     

A NOTE ON “COMMON DUE WINDOW SCHEDULING”

Kramer and Lee recently addressed a common due window scheduling problem with earliness and tardiness penalties, where earliness and tardiness penalty factors are constant and the common window size is given. They showed that the problem is polynomial when the location of the due window is a decision variable. For the case where the location of the due window is given, the problem is also polynomial when the latest due date is greater than or equal to the makespan, and they proposed a pseudopolynomial dynamic programming algorithm to find an optimal schedule when the latest due date is less than the makespan. In this note we address the problem for the case where the location of the due window is given. Specifically, we show that the problem is polynomial if the window location is unrestricted, and present a more efficient dynamic program algorithm to optimally solve the problem if the window location is restricted. The concepts of unrestricted and restricted window locations are defined in this note.     

Single-machine scheduling with production and rejection costs to minimize the maximum earliness

In this paper, we consider the single-machine scheduling problem with production and rejection costs to minimize the maximum earliness. If a job is accepted, then this job must be processed on the machine and a corresponding production cost needs be paid. If the job is rejected, then a corresponding rejection cost has to be paid. The objective is to minimize the sum of the maximum earliness of the accepted jobs, the total production cost of the accepted jobs and the total rejection cost of the rejected jobs. We show that this problem is equivalent to a single-machine scheduling problem to minimize the maximum earliness with two distinct rejection modes. In the latter problem, rejection cost might be negative in the rejection-award mode which is different from the traditional rejection-penalty mode in the previous literatures. We show that both of two problems are NP-hard in the ordinary sense and then provide two pseudo-polynomial-time algorithms to solve them. Finally, we also show that three special cases can be solved in polynomial time.     

Scheduling with release times and rejection on two parallel machines

In this paper we study scheduling with release times and job rejection on two parallel machines. In our scheduling model each job is either accepted and then processed by one of the two machines at or after its release time, or it is rejected and then a rejection penalty is paid. The objective is to minimize the makespan of the accepted job plus the total penalty of all rejected jobs. The scheduling problem is NP-hard in the ordinary sense. In this paper, we develop a \(1.5+\epsilon \)-approximation algorithm for the problem, where \(\epsilon \) is any given small positive constant.     

Single-machine multitasking scheduling with job efficiency promotion

Motivated by behavioural and psychological phenomena that occur in human operators, we study single-machine multitasking scheduling with job efficiency promotion. In traditional multitasking scheduling, the primary task is assumed to be interrupted by every waiting task. In this paper we take into account job efficiency promotion that helps reduce the actual interruption time. We propose two functions to model job efficiency promotion based on the job positions in a given schedule. The objective is to minimize the makespan, total completion time, and total absolute difference in completion times. We show that the problem is polynomially solvable for each objective. We also provide efficient solutions for some special cases.

     

Experimental earliness/tardiness production planning with the due window system

In order to use the philosophy of JIT to improve the production planning method of MRP-II, we propose the experimental software system of the earliness/tardiness produc tion planning problem with due window. By means of the approaches and model reported in this paper, the optimal production planning can be achieved. The recommended model extends the problem of due window from the shop scheduling level into the aggregated planning level of mass manufacturing systems. Simulation results have demonstrated that the experimental software is a useful tool for the production management of repetitive manufacturing enterprises.     

带工期指派的产品服务系统订单随机调度问题研究

针对由一个制造工厂和多个区域服务中心组成的服务型制造企业,研究了考虑生产时间和服务时间均具有随机性且工期可指派的产品服务系统（PSS）订单调度问题。首先以最小化订单提前、误工和工期指派费用的期望总额为目标构建问题的优化模型,然后分析目标函数近似值的最优性条件,据此提出加权最短平均生产时间排序规则,并结合该规则与插入邻域局部搜索设计了启发式算法对问题进行求解,最后通过数值仿真验证算法的可行性和有效性。研究表明,提前费用偏差对PSS订单调度与工期指派决策的影响很小,因此企业管理者无需准确估计库存费用也能制定出比较有效的PSS订单调度策略;而工期指派费用偏差对决策结果的影响非常大,因此企业管理者在决策时必须谨慎估计该项费用。     

Scheduling with machine cost and rejection

In this paper we consider the scheduling problem with machine cost and rejection penalties. For this problem, we are given a sequence of independent jobs, each being characterized by its processing time (size) and its penalty. No machine is initially provided, and when a job is revealed the algorithm has the option to purchase new machines. Right when a new job arrives, we have the following choices: (i) reject it, in which case we pay its penalty; (ii) non-preemptively process it on an existing machine, which contributes to the machine load; (iii) purchase a new machine, and assign it to this machine. The objective is to minimize the sum of the makespan, the cost for purchasing machines, and the total penalty of all rejected jobs. For the small job case, (where all jobs have sizes no greater than the cost for purchasing one machine, and which is the generalization of the Ski-Rental Problem) we present an optimal online algorithm with a competitive ratio of 2.     

Online scheduling with rejection and reordering: exact algorithms for unit size jobs

We study an online scheduling problem with rejection on \(m\ge 2\) identical machines, in which we deal with unit size jobs. Each arriving job has a rejection value (a rejection cost or penalty for minimization problems, and a rejection profit for maximization problems) associated with it. A buffer of size \(K\) is available to store \(K\) jobs. A job which is not stored in the buffer must be either assigned to a machine or rejected. Upon the arrival of a new job, the job can be stored in the buffer if there is a free slot (possibly created by evicting other jobs and assigning or rejecting every evicted job). At termination, the buffer must be emptied. We study four variants of the problem, as follows. We study the makespan minimization problem, where the goal is to minimize the sum of the makespan and the penalty of rejected jobs, and the \(\ell _p\) norm minimization problem, where the goal is to minimize the sum of the \(\ell _p\) norm of the vector of machine completion times and the penalty of rejected jobs. We also study two maximization problems, where the goal in the first version is to maximize the sum of the minimum machine load (the cover value of the machines) and the total rejection profit, and in the second version the goal is to maximize a function of the machine completion times (which measures the balance of machine loads) and the total rejection profit. We show that an optimal solution (an exact solution for the offline problem) can always be obtained in this environment, and determine the required buffer size. Specifically, for all four variants we present optimal algorithms with \(K=m-1\) and prove that in each case, using a buffer of size at most \(m-2\) does not allow the design of an optimal algorithm, which makes our algorithms optimal in this respect as well. The lower bounds hold even for the special case where the rejection value is equal for all input jobs.