Understanding how and why changes propagate during engineering design is critical because most products and systems emerge from predecessors and not through clean sheet design. This paper examines a large data set from industry including 41,500 change requests that were generated during the design of a complex sensor system spanning a period of 8 years. In particular, the networks of connected parent, child, and sibling changes are resolved over time and mapped to 46 subsystem areas of the sensor system. These change networks are then decomposed into one-, two-, and three-node motifs as the fundamental building blocks of change activity. A statistical analysis suggests that only about half (48.2%) of all proposed changes were actually implemented and that some motifs occur much more frequently than others. Furthermore, a set of indices is developed to help classify areas of the system as acceptors or reflectors of change and a normalized change propagation index shows the relative strength of each area on the absorber-multiplier spectrum between-1 and +1. Multipliers are good candidates for more focused change management. Another interesting finding is the quantitative confirmation of the "ripple" change pattern previously proposed. Unlike the earlier prediction, however, it was found that the peak of cyclical change activity occurred late in the program driven by rework discovered during systems integration and functional testing.

The sheer scale of large engineering projects is impressive in terms of the number of people involved, the number of components designed and the time-scale of the project. For example, an aircraft has tens of thousands of components designed and developed by teams of thousands of engineers at the main aircraft manufacturer in conjunction with a large supply chain over a period of about five to ten years. Major components such as the avionics, the undercarriage or the engine are sourced to main first-tier suppliers. These major components themselves are highly complex products with hundreds of components designed by a large team of people and long supply chains. For example, designers have commented in discussions that the design effort going.

In the automotive industry, engine companies produce a highly complex product, with a need to satisfy stringent legislation emission targets while at the same time they are faced with a highly fragmented market of different customers demanding highly customized products. A main driver to be competitive in such a situation is to manage changes effectively. Change requests from new customer requirements even late in the process must be analyzed for potential costly knock-on effects to other, not necessarily connected components. This paper describes how the Change Prediction Method (CPM) can be applied for the assessment of knock-on change risks which supports companies in planning for changes before being implemented and allows for an improved planning towards mass customization.

Managing design freezes plays an important part in today's competitive markets and successful freeze management can make the difference between delivering a product in time and in budget or failed design projects. On the one hand, design managers want to limit change propagation by freezing components or component-interfaces in order to avoid unwanted change propagation and to meet design lead-times. On the other hand freezing the design of a component too early can constrain innovation and having to unfreeze a design that was frozen before can add further redesign costs. This paper introduces an algorithm for determining an optimal change freeze order (CFO) based on combined change risks and component redesign cost, which has been applied in an automotive company in the design of a diesel engine.

Engineering change is a fundamental part of all design activities at all stages of design; complex products are designed by modification from existing ones; requirements change during long development projects; or problems through the design process or product use require rework. Yet engineering change has only recently become the topic of academic research and only few specific tools exist to manage engineering change. This paper reports on two workshops held in 2008, one in the UK and one in the US, where practising engineers reported on the challenges their organisations face with engineering changes and what approaches they take to deal with the effects of change. An analysis of 12 presentations shows that the causes for engineering change are very similar between different industries. However the approaches taken by companies to assess, manage and potentially take advantage of engineering changes are very different and in the absence of specific tools for managing engineering change companies used general process improvement and systems engineering tools, ranging from virtual design and QFD to high-level system simulation.

• Change propagation analysis based on DSM • Supporting the assessment of design strategies • Feasible for industrial application • Future work - supply chain and manufacturing cost.

When designing complex products such as robots or jet engines, companies face the problem that designers lack the necessary tools to predict the behaviour of the product in the case of component change, and to assess the risks associated with decisions. Product models allow companies and individual designers to reason about product properties. The information in the model can be used to analyse the properties of a product before decisions are taken about potential modifications. To meet time and budget constraints it is vital to have the ability to predict the risks of knock-on effects before a change is implemented and select alternative changes accordingly. This paper introduces a comparison of two product models used in the Change Prediction Method (CPM) and the Contact & Channel Model (C&CM). It analyses ways where both approaches can benefit and complement each other in analysing and predicting product change.

• Two component classification schemes based on components DSM were introduced • Provide information and strategies from a change perspective - PVP identifies components for change strategies - PAMP indicates components that could be frozen early • Help designers plan for new product variants • Both schemes can be used concurrently to provide more insights.

Adjacency matrices or DSMs (design structure matrices) and node-link diagrams are both visual representations of graphs, which are a common form of data in many disciplines. DSMs are used throughout the engineering community for various applications, such as process modelling or change prediction. However, outside this community, DSMs (and other matrix-based representations of graphs) are rarely applied and node-link diagrams are very popular. This paper will examine, which representation is more suitable for visualising graphs. For this purpose, several user experiments were conducted that aimed to answer this research question in the context of product models used, for example in engineering, but the results can be generalised to other applications. These experiments identify key factors on the readability of graph visualisations and confirm work on comparisons of different representations. This study widens the scope of readability comparisons between node-link and matrix-based representations by introducing new user tasks and replacing simulated, undirected graphs with directed ones employing real-world semantics.

Change to existing products is fundamental to design processes. New products are often designed through change or modification to existing products. Specific parts or subsystems are changed to similar ones whilst others are directly reused. Design by modification applies particularly to safety critical products where the reuse of existing working parts and subsystems can reduce cost and risk. However change is rarely a matter of just reusing or modifying parts. Changing one part can propagate through the entire design leading to costly rework or jeopardising the integrity of the whole product. This paper characterises product change based on studies in the aerospace and automotive industry and introduces tools to aid designers in understanding the potential effects of change. Two ways of supporting designers are described: probabilistic prediction of the effects of change and visualisation of change propagation through product connectivities. Change propagation has uncertainties which are amplified by the choices designers make in practice as they implement change. Change prediction and visualisation is discussed with reference to complexity in three areas of product development: the structural backcloth of connectivities in the existing product (and its processes), the descriptions of the product used in design and the actions taken to carry out changes.

This paper describes research on the use of multiple views for modelling products and processes in the design of complex products. Single visual representations of design models only provide a limited perspective, hiding important information from the designer. Based on an industrial case study with an engine company, the utility of multiple views in analysing model data is demonstrated. The case study showed an industrial need for improved visualisation techniques, as the currently used method - Gantt charts of design tasks - did not provide enough information in order to properly steer the design process and predict process behaviour. This approach of using a variety of different visualisation techniques together with improved techniques for modelling and simulating design processes provides insight into the hidden dependencies between the design artefact (the product) and its design process.

Reasoning about the connectivity within a product is an integral part of many core design activities. Due to the complexity of a product and the sheer number of potential links, designers often overlook vital connections resulting in problems later in the process, leading to errors or costly rework. Product connectivity models, which are essentially graphs, are a promising approach for capturing these links between components in a complex product. The primary visual representation used to create such connectivity models is the Design Structure Matrix (DSM). However, other representations of graphs may be superior for creating connectivity models of products. This paper presents node-link displays as equally valid representations for product connectivity models and reports on an experimental study that investigates whether DSMs or node-link diagrams are more suitable for building such models.

Planning the engineering design phase of NPD projects is non-trivial due to uncertainty from a number of sources. This research uses simulation of process models to explore project sensitivity to rework. Our results highlight the importance of iteration due to task failure by showing that different aspects of rework behavior have a major impact on project duration uncertainty