On the Construction of Engineering Handbooks : With an Illustration From the Railway Safety Domain

This book focuses on the clarification of what actually a handbook is, the systematic identification of what ought to be considered as “settled knowledge” (extracted from historic repositories) for inclusion into such a handbook, and the “assembly” of such identified knowledge into a form which is fit for the purpose and conforms to the formal characteristics of handbooks as a “literary genre”. For many newly emerging domains or disciplines, for which no handbook with normative authority has yet been defined, the question arises of how to do this systematically and in a non-arbitrary manner. This book is the first to reflect upon the question of how to construct a desktop handbook. It is demonstrated how concept analysis can be used for identifying settled knowledge as the key ingredient by utilizing the assembled data for classification; a presentation scheme for handbook articles is developed and demonstrated to be suitable. The sketched approach is then illustrated by an example from the railway safety domain. Finally, the limitations of the presented methods are discussed. The key contribution of this book is the (example illustrated) construction method itself, not the handbook, which would result from a highly detailed and thoroughly comprehensive application of the method. Foreword 6 Preface 8 Acknowledgements 10 Contents 11 Part I Background 13 Chapter 1: Introduction and Motivation 14 1.1 What constitutes a HB? 16 1.2 Settled Knowledge 19 1.3 Formal Methods in Design and Validation of Railway Control Systems 20 1.4 Structure of the Remainder of this Book 21 Chapter 2: Related Work 23 2.1 Organisation of Engineering Knowledge 23 2.2 Other Approaches to HB Construction 25 2.3 Other Classifications of Formal Methods 27 2.4 Comparison with our Approach 29 Part II Analysis 31 Chapter 3: A General Method for Composing an Engineering HB 32 3.1 Step 1: Choice of Sources for Settled Knowledge 36 3.2 Step 2: Data Collection and Choice of Domain 39 3.3 Step 3: Application of FCA to the Data Collected 41 3.4 Step 4: Choice of a Stability Threshold 44 3.5 Step 5: Classification of Settled Knowledge 46 3.6 Step 6: Presentation of Settled Knowledge 48 Chapter 4: Application of the General Method to the Railway Domain 50 4.1 Step 1: Choice of Sources for the Railway Domain 51 4.2 Step 2: Data Collection and Choice of Domain Specific Attributes for FCA 51 4.3 Step 3: Application of FCA to the Data Collected 52 4.4 Step 4: Choice of a Stability Threshold 54 4.5 Step 5: Interpretation and Discussion of the Detected Settled Knowledge 56 4.5.1 Observations (and Peculiarities) 56 4.5.2 Settledness 58 4.5.3 Limitations of our Findings 58 4.5.4 Guidance for a Handbook on FMs for the Railway Domain 59 4.6 Possible Threats to Validity: Critical Evaluation of Steps 1–5 in the Context of our Railway Example 59 4.6.1 Notion of ‘Settledness’ 60 4.6.2 Choice of Database 60 4.6.3 Choice of Formal Concept Analysis 61 4.6.4 Choice of Attributes for FCA 61 4.6.5 Choice of Stability Threshold 61 4.6.6 Classification of Settled Knowledge 62 Part III Synthesis 63 Chapter 5: Example HB Entry of a Formal Method for the Railway Domain — Step 6 64 5.1 Problem Class: Verification through Model Checking of Ladder Logic Programs for Safety 65 5.2 Solution 66 5.3 Academic Explanation 76 5.3.1 A Short Bibliography 76 5.3.2 Solid State Interlockings 76 5.3.3 Explanation for (Step 1): Tseitin Transformation 78 5.3.4 Explanation for (Step 2): Discrete Time, Temporal First Order Logics 79 5.3.5 Explanation of Step (3): Verification Problem associated with Ladder Logic 80 5.4 Experience Reports Concerning Step 6 81 Chapter 6: Conclusions and Prospects for Future Work 83 References 85