A methodological approach for the Product Development Process optimization of aircraft components

The way the product development process (PDP) is executed in industry is changing over time. Over the years, a paradigm shift can be observed, where starting from a purely technical process, the PDP is now integrating both technical and business process aspects. The overall process performance is becoming increasingly important. This has resulted in product development techniques like concurrent engineering and lean product development principles. The goal of the PDP is to deliver a product that meets all technical requirements, while keeping lead time and development cost as low as possible. Maximizing technical performance while minimising development costs and lead time is a great challenge, especially in case of complex product such as aircraft and aircraft systems. Here the number of interacting systems and involved disciplines is large, and the type and strength of such interaction is not always easy to grasp and manage effectively. For many companies, lead time is the most important performance measure of their development process, because their competitive edge on the market depends on it. A reduced time-to-market results in a reduction in cost-of-delay and a larger market share [2]. The challenge is to keep track of the performance of the product and process simultaneously, and be able to adjust the process on time reacting effectively and efficiently to changes. This becomes even more challenging if multiple disciplines and development locations need to be managed. This paper describes the implementation of a proposed methodology to optimize the PDP process of an aircraft rudder, accounting for different criteria such as lead-time, recurring and non-recurring costs, risk and combinations thereof. Two use cases are taken into account: (1) optimisation of a known process prior to the start of the process;(2) optimisation in the course of a project, in which tasks might have progressed and/or resources allocation modified, (e.g. another project demands attention of all stress engineers), and/or customer requirements changed (e.g. a different load spectrum needs to be used). The rudder development process is representative for a typical complex product PDP, where both manual and automated tasks are present and multiple departments (e.g. stress-, design-, manufacturing and cost engineering) are involved. The development process of a rudder also is highly iterative. An example can be found in the design of a hinge of a rudder, here design starts often at the centerline of the hinge (e.g. a bolt or pin) and the development team works outward towards the rudder outer mold line, which constraints the hinge contour. During this process multiple disciplines are taken into account to assess strength, cost, manufacturability, material combinations etc. These disciplines influence each other and hence a change in for example the material of the pin can leads to a redesign of the diameter based on stress calculations. The proposed implementation makes use of a Workflow Management System (WfMS) with an integrated optimisation tool to give the user the possibility to formalize and optimally reconfigure a certain PDP flow, accounting for the two use cases above. The paper illustrates the possible performance optimization of the rudder PDP, by changing the sequencing of tasks and their resources allocation, taking into account the current status of the PDP (e.g. progress of tasks, fixed parameters etc.) and the available resources. To be able to perform this optimisation, DSM partitioning will be used to assess the possibilities of re- sequencing, parallelisation of tasks, reducing feedback loops. Figure 1 shows the strength of mapping a workflow (a) on a DSM (b) followed by partitioning of the DSM (c). Also the opportunities of overlapping tasks with low product data sensitivity will be assessed by means of numerical partitioning algorithms [1]. The DSM model is used to optimise process sequence for lead time and modularity. E.g. minimising the impact of changes on the total lead time by minimising the feedback loops in the DSM. Also the clustering of tasks in the DSM would allow for effective process innovation, targeted by innovating the clustered tasks by application of KBE automation. By mimicking the automation of a task the effect on leadtime of the total process can be assessed.