James Little

From enterprise models to scheduling models: bridging the gap

Enterprise models cover all aspects of modern enterprises, from accounting, through management of custom orders and invoicing, to operational data such as records on machines and workers. In other words, all data necessary for running the company are available in enterprise models. However, these data are not in the proper format for some tasks such as scheduling and optimization. Namely, the concepts and terminology used in enterprise models are different from what is traditionally used in scheduling and optimization software. This paper deals with the automated translation of data from the enterprise model to a scheduling model and back. In particular, we describe how to extract data from the enterprise model for solving the scheduling problem using constraint-based solvers.

On supporting Lean methodologies using constraint-based scheduling

Lean Manufacturing—often simply referred to as “Lean”—is a process management philosophy that aims to improve the way in which products are manufactured. It does this through identifying and removing waste and creating a smooth transition between stages in the production process. To a large extent, it relies on visual and simple mechanical aids to assist in improving manufacturing effectiveness. However, when it comes to combining several aspects of Lean or when dealing with complex environments, quantitative modelling becomes essential to achieve the full benefits of Lean.In this paper, we show through two detailed case studies how various aspects of Lean can be supported using (constraint-based) scheduling tools. One study concerns a planning support tool to evaluate different Lean initiatives; the other supports the day-to-day scheduling of a complex, Leaned production process

An integrated business rules and constraints approach to data centre capacity management

A recurring problem in data centres is that the constantly changing workload is not proportionally distributed over the available servers. Some resources may lay idle while others are pushed to the limits of their capacity. This in turn leads to decreased response times on the overloaded servers, a situation that the data centre provider wants to prevent. To solve this problem, an administrator may move (reallocate) applications or even entire virtual servers around in order to spread the load. Since there is a cost associated with moving applications (in the form of down time during the move, for example), we are interested in solutions with minimal changes. This paper describes a hybrid approach to solving such resource reallocation problems in data centres, where two technologies have to work closely together to solve this problem in an efficient manner. The first technology is a Business Rules Management System (BRMS), which is used to identify which systems are considered to be overloaded on a systematic basis. Data centres use complex rules to track the behaviour of the servers over time, in order to properly identify overloads. Representing these tracking conditions is what the BRMS is good for. It defines the relationships (business constraints) over time between different applications, processes and required resources that are specific to the data centre. As such, it also allows a high degree of customisation. Having identified which servers require reallocation of their processes, the BRMS then automatically creates an optimisation model solved with a Constraint Programming (CP) approach. A CP solver finds a feasible or the optimal solution to this CSP, which is used to provide recommendations on which workload should be moved and whereto. Notice that our use of a hybrid approach is a requirement, not a feature: employing only rules we would not be able to compute an optimal solution; using only CP we would not be able to specify the complex identification rules without hard-coding them into the program. Moreover, the dedicated rule engine allows us to process the large amounts of data rapidly.

CP-INSIDE: Embedding Constraint-Based Decision Engines in Business Applications

The CP-INSIDE project seeks to make constraint programming technology more accessible. In particular, it aims to facilitate the embedding of CP technology in business applications, and its integration with familiar software tools. A major objective of the CP-INSIDE project is to allow business developers to create constraint-based decision engines utilizing the power of CP technology in a vendor-neutral way. CP-INSIDE simplifies the development of specialized decision engines making them independent of the underlying generic CP solvers. There have already been several attempts to unify constraint programming languages (for instance, [1,2,3]) that directly or indirectly contributed to the same objective.

Thermal test scheduling using constraint programming

Abstract Temperature cycling test is one of the key stages in the process of testing circuit packs in telecommunications. To obtain a good overall test schedule requires that the thermal test is carried out efficiently i.e. with the minimum number of runs and valid configurations of packs at each run. However, finding valid configurations and building them into a minimal thermal test schedule is a difficult combinatorial problem. Constraint Programming allows both a way of modelling the rules of configuration and formulating a model to derive an optimal number of runs. We describe this model and the results obtained from it for a large multi-national telecommunications manufacturer.

Optimising Machine Selection Rules for Sequence Dependent Setups with an Application to Cartoning

Abstract Conventional scheduling technology, which tries to optimise performance metrics such as utilisation and makespan, works well in environments where there is a high degree of stability, and hence certainty. However, in uncertain situations, schedules that try to achieve optimality have trouble achieving this target. An often used technique to circumvent the issues with uncertainty that optimal policies display, is the use of rules that are triggered when a job enters the system or a machine becomes available. Such rules are by definition reactive, and can thus deal very well with uncertainty. The downside of these rules is that they make local decisions, which may result in non-optimal behaviour. This becomes especially apparent in light of significant sequence dependent setup times. These are, by nature, dependent upon the global sequence, something that rules cannot deal well with. In this paper, we investigate a method to generate custom machine selection rules that lead to improved setup times. We illustrate the advantages of this new type of rule by presenting an experimental analysis of using these rules at the cartoning department of a large manufacturing company.

Scheduling for cellular manufacturing

Alcatel-Lucent is a major player in the field of telecommunications. One of the products it offers to network operators is wireless infrastructure such as base stations. Such equipment is delivered in cabinets. These cabinets are packed with various pieces of electronics: filters, amplifiers, circuit packs, etc. The exact configuration of a cabinet is dependent upon the circumstances it is being placed in, and some 20 product groups can be distinguished. However, the variation in cabinets is large, even within one product group. For this reason, they are built to order. In order to improve cost, yield and delivery performance, lean manufacturing concepts were applied to change the layout of the factory to one based on cells. These cells focus on improving manufacturing through standardised work, limited changeovers between product groups and better utilisation of test equipment. A key component in the implementation of these improvements is a system which schedules the cells to satisfy customer request dates in an efficient sequence. This paper describes the transformation and the tool that was built to support the new method of operations. The implementation has achieved significant improvements in manufacturing interval, work in process inventory, first test yield, headcount, quality (i.e. fewer defects are found during the testing stage) and delivery performance. Although these benefits are mainly achieved because of the change to a cell layout, the scheduling tool is crucial in realising the full potential of it.

PROCESS DESIGN FOR EFFICIENT SCHEDULING

Abstract For manufacturing, management needs to make rapid, informed strategic decisions to react to changes in the market place. In many of these cases the decisions will give rise to a redesign of the manufacturing processes (layout, changes in product mix and volumes), which in turn changes the nature of the scheduling problems associated with the production facilities. However, the necessary skills and time in bridging the gap between design and scheduling are not available to allow managers to evaluate the many different high level decisions. In this paper, we propose an architecture which brings some scheduling capabilities to managers and planners. This approach is evaluated on two real-life design scenarios considered by a manufacturing company.