Wout van Wezel

The SEC-system reuse: support for scheduling system development

Recently, in a joint cooperation of Stichting VNA, SAL Apotheken, the Faculty of Management and Organization, and the University Centre for Pharmacy, University of Groningen in the Netherlands, a Ph.D-study started regarding Apot(he)ek, Organization and Management (APOM). The APOM-project deals with the structuring and steering of pharmacy organization. The manageability of the internal pharmacy organization, and the manageability of the direct environment of pharmacy organization is the subject matter. The theoretical background of the APOM-project is described. A literature study was made to find mixes of objectives. Three mixes of objectives in pharmacy organization are postulated; the product mix, the process mix, and the customer mix. The typology will be used as a basic starting point for the empirical study in the next phase of the APOM-project.

Allocating functions to human and algorithm in scheduling

An important part of Advanced Planning Systems are algorithms. When algorithms are applied, the task is automated as much as possible. However, the human that is supposed to use the algorithm is generally ignored during the development process. As a consequence, a prior investigation whether and how an algorithm can or will be used in practice is not integrated in the development process. In contrast, in the field of cognitive ergonomics, function allocation methods explicitly take into account human factors in the design of human/computer systems. The function allocation literature, however, is mainly focused on dynamic systems where humans must make decisions in situations with time pressure and important safety risks, e.g., nuclear plants and air traffic control. We analyze the differences between such dynamic systems and planning and scheduling, and we propose a model for function allocation in planning and scheduling taking into account cognitive and human-machine cooperation aspects.

A Comparison of Task Analysis Methods for Planning and Scheduling

Planning and scheduling experts in practice are often faced with the question of how a company can improve its planning performance. Such improvements can be related to, for example, computer support, organizational task division, performance analysis, etc. The multitude of planning and scheduling factors and their interrelatedness makes it difficult to integrally explain current performance and assess the consequences of changes. We analyze how different perspective on task analysis methods complement each other for the various questions that planning and scheduling experts encounter in practice. There are two main findings. On the one hand, a combination of methods is often necessary in order to avoid myopia and biased results. On the other hand, however, the analysis shows that not all questions require a full-scale analysis of the situation.

Cognition, tasks and planning: supporting the planning of shunting operations at the Netherlands Railways

One of the issues in planning research and planning practice are their apparent incompatibility. Various fields study aspects of planning, but too little is jointly used in practice. Starting from the human planner as a cognitive system, we elaborate the mixed initiative approach, in which we combine three fields: cognitive aspects of the task of the human planner, computer support, and algorithms. In this article, we argue that algorithms should be created for planners’ subtasks rather than for planning problems. We demonstrate this in a prototype that we developed for a shunting planning project we are involved in at the Netherlands Railways [Nederlandse Spoorwegen (NS)]. In this project, a task analysis resulted in a planning support system with bottom up designed scheduling algorithms.

Revisiting Decision Support Systems for Cognitive Readiness: A Contribution to Unstructured and Complex Scheduling Situations

In this article, the authors focus on scheduling situations. Because of their unstructured nature and hard combinatorial complexity, scheduling situations have always been a predominant application area for decision support systems (DSSes). After setting out the generic characteristics of a DSS, the authors summarize some of their known limitations in scheduling situations. They argue that scheduling situations focus excessively on performance and effectiveness and neglect the potential of human schedulers. The authors then review field studies to understand the way in which humans perform scheduling, demonstrating the importance of cognitive readiness if schedulers are to schedule efficiently. On the basis of design issues and field studies, the authors go on to refine three factors that influence cognitive readiness that therefore need to be taken into account when designing a DSS, namely, adaptability, flexibility, and acceptability. Finally, the authors show how these concepts can be used to assess DSS user interfaces and algorithms from a cognitive readiness perspective.

Design of Scheduling Algorithms

The accomplishment of a manufacturing company’s objectives is strongly connected to the efficient solution of scheduling problems that are faced in the production environment. Numerous methods for the solution of these problems have been published. However, very few of them have been adopted by manufacturing companies. This chapter suggests that the basic reason behind this imbalance is the inadequate representation of the scheduling process when designing decision support systems. Hence, the algorithms that are designed and included in these systems might not reflect the problems that actually have to be solved. The relevance of algorithmic design can be improved by using a more complete representation of the scheduling process, which would be highly relevant for increasing the adoption rate of new support systems.

Planning, anticipatory systems, and kinds of actors

Anticipation and planning have much in common. Anticipation means being ahead of things. Planning means not only anticipating on things to come, but also trying to determine a course of action to reach a goal. Planning is investigated in many domains, for example, in economy, biology, and in the cognitive sciences. Each of these domains uses its own research methodologies, languages, ontologies, and models. Although there are many differences between the various approaches and models, there must of course also be similarities, since they all deal with establishing a future course of actions to reach a goal. We will compare several planning approaches on a number of aspects, e.g., the kind of entity that makes the plan (natural or artificial), whether the plan is executed by the planner or by another entity, representational issues, and the domain that the plan is executed in. Our article focuses on analytical dimensions in the extended field of planning. We do not discuss empirical results.

Case Study: Advanced Decision Support for Train Shunting Scheduling

Train shunting scheduling for the Netherlands Railways is a complex planning problem that is performed by about 130 full-time planners. It concerns the planning of day-to-day shunting operations at the large stations in the railway network. Input for this planning is based on the long term schedule of arrivals and departures at a station, rolling stock and track maintenance schedules, and specific circumstances that have to be taken into account when preparing the plan for a particular day or night. Currently, the planners mainly work manually. The plans are made and revised on paper first, after which the outcome is put in the computer. In our study, the central question is: how can these planners be supported in their task with an advanced planning system?

Design of Scheduling Algorithms: Applications

This chapter discusses the insights developed for designing scheduling algorithms according to three design projects where algorithms have been developed. The choice of applications covers a broad spectrum. The methods used are from three different fields, namely combinatorial optimization, genetic (evolutionary) algorithms, and mathematical optimization. The application areas differ also in terms of the role of a human user of the algorithm. Some of these algorithms have been developed without detailed study of the competences of the perceived users. Others have examined humans when performing the scheduling tasks manually, but have not considered the change in cognitive load if the process of planning changes due to the new algorithm and computerized support. Although none of the design projects fulfils all criteria developed in the framework of Chap. 12, we show that the framework helps to assess the design projects and the resulting algorithms, and to identify the main weaknesses in these applications. Finally, we show how they can be addressed in future.

The Interconnectivity of Planning and Shop Floor: Case Description and Relocation Analysis

This chapter provides a detailed description of the planning situation within a manufacturer of office furniture. The description provides much insight into the reality of planning and rescheduling, including the dynamic relation between planning and the shop floor. A quasi-experiment is reported in which the planning department has been relocated closer to the shop floor. Expectations before and experiences after the relocation are measured and reported. Although no significant changes in communication behavior are observed, both planners and shop floor foremen perceived positive consequences of the relocation.