Report: Historical Exploration: Learning Lessons from the Past to Inform the Future

Full Report (revised link)

Tim Bedford, John Quigley, Matthew Revie and Lesley Walls Department of Management Science, Strathclyde Business School, 40, George St, Glasgow, G1 1QE, Scotland, UK.

Executive Summary

This report examines a number of exploration campaigns that have taken place during the last 700 years, and considers them from a risk perspective. The explorations are those led by Christopher Columbus, Sir Walter Raleigh, John Franklin, Sir Ernest Shackleton, the Company of Scotland to Darien and the Apollo project undertaken by NASA. To provide a wider context for investigating the selected exploration campaigns, we seek ways of finding analogies at mission, programmatic and strategic levels and thereby to develop common themes. Ultimately, the purpose of the study is to understand how risk has shaped past explorations, in order to learn lessons for the future. From this, we begin to identify and develop tools for assessing strategic risk in future explorations.

Figure 0.1 (see Page 6) summarizes the key inputs used to shape the study, the process and the results, and provides a graphical overview of the methodology used in the project. The first step was to identify the potential cases that could be assessed and to create criteria for selection. These criteria were collaboratively developed through discussion with a Business Historian. From this, six cases were identified as meeting our key criteria. Preliminary analysis of two of the cases allowed us to develop an evaluation framework that was used across all six cases to ensure consistency. This framework was revised and developed further as all six cases were analyzed.

A narrative and summary statistics were created for each exploration case studied, in addition to a method for visualizing the important dimensions that capture major events. These Risk Experience Diagrams illustrate how the realizations of events, linked to different types of risks, have influenced the historical development of each exploration campaign. From these diagrams, we can begin to compare risks across each of the cases using a common framework. In addition, exploration risks were classified in terms of mission, program and strategic risks. From this, a Venn diagram and Belief Network were developed to identify how different exploration risks interacted. These diagrams allow us to quickly view the key risk drivers and their interactions in each of the historical cases.

By looking at the context in which individual missions take place we have been able to observe the dynamics within an exploration campaign, and gain an understanding of how these interact with influences from stakeholders and competitors. A qualitative model has been created to capture how these factors interact, and are further challenged by unwanted events such as mission failures and competitor successes. This Dynamic Systemic Risk Model is generic and applies broadly to all the exploration ventures studied. This model is an amalgamation of a System Dynamics model, hence incorporating the natural feedback loops within each exploration mission, and a risk model, in order to ensure that the unforeseen events that may occur can be incorporated into the modeling.

Finally, an overview is given of the motivational drivers and summaries are presented of the overall costs borne in each exploration venture. An important observation is that all the cases – with the exception of Apollo – were failures in terms of meeting their original objectives. However, despite this, several were strategic successes and indeed changed goals as needed in an entrepreneurial way. The Risk Experience Diagrams developed for each case were used to quantitatively assess which risks were realized most often during our case studies and to draw comparisons at mission, program and strategic levels. In addition, using the Risk Experience Diagrams and the narrative of each case, specific lessons for future exploration were identified.

There are three key conclusions to this study:

Analyses of historical cases have shown that there exists a set of generic risk classes.

This set of risk classes cover mission, program and strategic levels, and includes all the risks encountered in the cases studied. At mission level these are Leadership Decisions, Internal Events and External Events; at program level these are Lack of Learning, Resourcing and Mission Failure; at Strategic Level they are Programmatic Failure, Stakeholder Perception and Goal Change. In addition there are two further risks that impact at all levels: Self-Interest of Actors, and False Model.

There is no reason to believe that these risk classes will not be applicable to future exploration and colonization campaigns.

We have deliberately selected a range of different exploration and colonization campaigns, taking place between the 15th Century and the 20th Century. The generic risk framework is able to describe the significant types of risk for these missions. Furthermore, many of these risks relate to how human beings interact and learn lessons to guide their future behavior. Although we are better schooled than our forebears and are technically further advanced, there is no reason to think we are fundamentally better at identifying, prioritizing and controlling these classes of risk.

Modern risk modeling techniques are capable of addressing mission and program risk but are not as well suited to strategic risk.

We have observed that strategic risks are prevalent throughout historic exploration and colonization campaigns. However, systematic approaches do not exist at the moment to analyze such risks.

A risk-informed approach to understanding what happened in the past helps us guard against the danger of assuming that those events were inevitable, and highlights those chance events that produced the history that the world experienced. In turn, it allows us to learn more clearly from the past about the way our modern risk modeling techniques might help us to manage the future – and also bring to light those areas where they may not.

This study has been retrospective. Based on this analysis, the potential for developing the work in a prospective way by applying the risk models to future campaigns is discussed. Follow on work from this study will focus on creating a portfolio of tools for assessing strategic and programmatic risk.