University of Twente Student Theses


Insights into the importance of scientific methods for tactical multi-project planning at AWL

Evers, Jeroen (2013) Insights into the importance of scientific methods for tactical multi-project planning at AWL.

[img] PDF
Abstract:AWL-Techniek is a company that builds automated welding machines, mainly for production lines in the automotive industry. AWL builds the machines in projects, each project has its own characteristics, since requirements are often tailored to the customer. Various departments with a limited amount of worker hours available are necessary to complete the projects. The motive for this research is that the capacity requirements of various capacity groups fluctuate strongly over weeks. This causes unnecessary high costs for the arrangement of non-regular capacity (overtime, temporary personnel, or outsourcing), a lower average utilisation of regular capacity, and busy periods in demand peaks. During these busy periods internal deadlines are often not met, which makes the planning and scheduling of successive activities and projects difficult. The objectives of this research are:  To give insight into the current planning methodology.  To advise on the applicability of recently developed methods and algorithms for planning, from scientific literature.  To show the benefit of algorithms for planning at AWL.  To make recommendations for the implementation of a new planning method. In this research we focus on tactical planning, which is the planning of high-level activities in order to be able to estimate capacity requirements for the coming months. These tactical plans are input for operational scheduling and define when activities can start and should be finished. Currently, the tactical planner makes a project plan when a new project is acquired without the help of an algorithm and without re-planning active projects. This research shows that AWL can benefit from planning from a multi-project planning approach. Examples in this report show that it is practically impossible to plan multiple projects with multiple activities efficiently by hand, because of the complexity of the planning problem. In this research we define an efficient tactical multi-project plan as a plan that, given all restrictions, is a valid plan with a low amount of non-regular hours needed and a low degree of fluctuation in capacity requirements over time. This report considers several approaches from the scientific literature. We extend one approach (ICPA) with a few steps, in order to make it suitable to cope with non-regular capacity restrictions. To our knowledge, this is the first non LP-based heuristic that is able to do this. This extension of ICPA is referred to in this report as Augmented ICPA. We perform an experiment, in which we re-plan the activities of 18 projects. The results show that, for the considered problem instance, a saving of 45% on the cost of non-regular capacity required is reached if we compare the current tactical plan to the plan resulting from the best performing algorithm. We argue that AWL can obtain a saving of between 0.6 and 1 million € on a yearly basis by employing an algorithm for tactical planning (depending on the intake of projects, input for- and effectiveness of the algorithm, and AWL’s ability to adhere to the resulting tactical plans). Other benefits of a new planning method are:  Less time spent on familiarising non-regular personnel with AWL.  Less time spent on planning.  More flexible planning.  Insight into the effects of (possible) strategic and tactical decisions.  Higher utilisation of regular capacity. In order to achieve the benefits of multi-project planning with an algorithm, AWL must take some steps:  First of all, AWL should measure a set of KPIs (proposed in this report) to able to say something about the performance of operations and of a (tactical) plan.  AWL should eliminate ambiguity in their data and consequently adhere to strict rules, in order to be able to provide standardised and complete input data for a new planning system. Certain data needed for the proper employment of an algorithm is currently not available at AWL. An important element is the maximum non-regular capacity restrictions of the various capacity groups.  In order to be able to plan from a multi-project perspective, a central tactical planning function should be responsible for the tactical project plan, also during the execution of the project. This way, the central planning function can manage the portfolio of projects and efficiently allocate capacity to project activities to guard the interests of AWL as a whole. When AWL completes these steps, some experienced advanced planning software providers should be contacted and the possibilities of building a planning package tailored to AWL should be investigated. It is important to create support and trust in a new planning system; this report also provides some recommendations to this end.
Item Type:Essay (Master)
AWL Techniek B.V.
Faculty:BMS: Behavioural, Management and Social Sciences
Subject:85 business administration, organizational science
Programme:Industrial Engineering and Management MSc (60029)
Link to this item:
Export this item as:BibTeX
HTML Citation
Reference Manager


Repository Staff Only: item control page