Christoph H. Loch

On Uncertainty, Ambiguity, and Complexity in Project Management

This article develops a model of a project as a payoff function that depends on the state of the world and the choice of a sequence of actions. A causal mapping, which may be incompletely known by the project team, represents the impact of possible actions on the states of the world. An underlying probability space represents available information about the state of the world. Interactions among actions and states of the world determine the complexity of the payoff function. Activities are endogenous, in that they are the result of a policy that maximizes the expected project payoff.A key concept is theadequacy of the available information about states of the world and action effects. We express uncertainty, ambiguity, and complexity in terms of information adequacy. We identify three fundamental project management strategies: instructionism, learning, and selectionism. We show that classic project management methods emphasize adequate information and instructionism, and demonstrate how modern methods fit into the three fundamental strategies. The appropriate strategy is contingent on the type of uncertainty present and the complexity of the project payoff function. Our model establishes a rigorous language that allows the project manager to judge the adequacy of the available project information at the outset, choose an appropriate combination of strategies, and set a supporting project infrastructure--that is, systems for planning, coordination and incentives, and monitoring.

Managing project uncertainty: from variation to chaos

This publication contains reprint articles for which IEEE does not hold copyright. Full text is not available on IEEE Xplore for these articles.

Selectionism and Learning in Projects with Complexity and Unforeseeable Uncertainty

Companies innovating in dynamic environments face the combined challenge of unforeseeable uncertainty (the inability to recognize the relevant influence variables and their functional relationships; thus, events and actions cannot be planned ahead of time) and high complexity (large number of variables and interactions; this leads to difficulty in assessing optimal actions beforehand).There are two fundamental strategies to manage innovation with unforeseeable uncertainty and complexity: trial and error learning and selectionism. Trial and error learning involves a flexible (unplanned) adjustment of the considered actions and targets to new information about the relevant environment as it emerges. Selectionism involves pursuing several approaches independently of one another and picking the best one ex post.Neither strategy nor project management literatures have compared the relative advantages of the two approaches in the presence of unforeseeable uncertainty and complexity. We build a model of a complex project with unforeseeable uncertainty, simulating problem solving as a local search on a rugged landscape. We compare the project payoff performance under trial and error learning and selectionism, based on a priori identifiable project characteristics: whether unforeseeable uncertainty is present, how high the complexity is, and how much trial and error learning and parallel trials cost. We find that if unforeseeable uncertainty is present and the team cannot run trials in a realistic user environment (indicating the projects true market performance), trial and error learning is preferred over selectionism. Moreover, the presence of unforeseeable uncertainty can reverse an established result from computational optimization: Without unforeseeable uncertainty, the optimal number of parallel trials increases in complexity. But with unforeseeable uncertainty, the optimal number of trials might decrease because the unforeseeable factors make the trials less and less informative as complexity grows.

Dynamic Portfolio Selection of NPD Programs Using Marginal Returns

Selecting program portfolios within a budget constraint is an important challenge in the management of new product development (NPD). Optimal portfolios are difficult to define because of the combinatorial complexity of project combinations. However, at the aggregate level of the strategic allocation of resources across product lines, investment in a program is not an all-or-nothing decision, but can beadjusted, resulting in a higher or lower program benefit (e.g., higher or lower quality). In some cases, resources can be adjusted even for individual projects.With this insight, one can usemarginal analysis to optimally allocate the scarce budget. This article develops a dynamic model of resource allocation, taking into account multiple interacting factors, such as independent or correlated, uncertain market payoffs that change over time, increasing or decreasing returns from the NPD investment, carry-over of the investment benefit over multiple periods, and interactions across market segments. We characterize optimal policies in closed form and derive qualitative decision rules for managers.

Social Preferences and Supply Chain Performance: An Experimental Study

Supply chain contracting literature has traditionally focused on aligning incentives for economically rational players. Recent work has hypothesized that social preferences, as distinct from economic incentives, may influence behavior in supply chain transactions. Social preferences refer to intrinsic concerns for the other partys welfare, reciprocating a history of a positive relationship, and intrinsic desires for a higher relative payoff compared with the other partys when status is salient. This article provides experimental evidence that social preferences systematically affect economic decision making in supply chain transactions. Specifically, supply chain parties deviate from the predictions provided by self-interested profit-maximization models, such that relationship preference promotes cooperation, individual performance, and high system efficiency, sustainable over time; whereas status preference induces tough actions and reduces both system efficiency and individual performance.

Parallel and Sequential Testing of Design Alternatives

An important managerial problem in product design in the extent to which testing activities are carried out in parallel or in series. Parallel testing has the advantage of proceeding more rapidly than serial testing but does not take advantage of the potential for learning between tests, thus resulting in a larger number of tests. We model this trade-off in the form of a dynamic program and derive the optimal testing strategy (or mix of parallel and serial testing) that minimizes both the total cost and time of testing. We derive the optimal testing strategy as a function of testing cost, prior knowledge, and testing lead time. Using information theory to measure the test efficiency, we further show that in the case of imperfect testing (due to noise or simulated test conditions), the attractiveness of parallel strategies decreases. Finally, we analyze the relationship between testing strategies and the structure of design hierarchy. We show that a key benefit of modular product architecture lies in the reduction of testing cost.

Managing the Process of Engineering Change Orders: The Case of the Climate Control System in Automobile Development

Abstract Engineering change orders (ECOs) are part of almost every development process, consuming a significant part of engineering capacity and contributing heavily to development and tool costs. Many companies use a support process to administer ECOs, which fundamentally determines ECO costs. This administrative process encompasses the emergence of a change (e.g., a problem or a market-driven feature change), the management approval of the change, up to the change’s final implementation. Despite the tremendous time pressure in development projects in general and in the ECO process in particular, this process can consume several weeks, several months, and in extreme cases even over 1 year. Based on an in-depth case study of the climate control system development in a vehicle, we identify five key contributors to long ECO lead times: a complex approval process, snowballing changes, scarce capacity and congestion, setups and batching, and organizational issues. Based on the case observations, we outline a number of improvement strategies an organization can follow to reduce its ECO lead times.

Evaluating growth options as sources of value for pharmaceutical research projects

The financial value of research projects is difficult to assess because they are highly uncertain. Often, the result is either an overly conservative approach to strategic innovation, based on net present value analyses, or an overly aggressive approach based on optimistic qualitative portfolios. R&D project evaluation requires recognizing threats as well as opportunities from uncertain events, and incorporating flexibility in managerial action in response to them. Real options pricing analysis is a widely discussed tool for evaluating such managerial flexibility. The limitation of options pricing lies in its requirement for complete financial markets, in which a replicating asset can be found that reproduces (or, at least, is correlated with) the project’s payoffs in all possible states of the world. However, the major risks of research projects are typically project specific and cannot be replicated in external markets. In this situation, a decision tree is a better tool to represent managerial options during execution of the project, and to evaluate its value. A decision tree is equivalent to options pricing for risks that can be priced in the financial markets (if trading of securities is explicitly included), and moreover, it can incorporate risks and flexibility that are not traded in financial markets. Using decision trees, we demonstrate a quantitative evaluation of compound growth options from research at BestPharma, a large international pharmaceutical company. A growth option is a future opportunity that may arise from a current R&D investment. The growth option may not be related to the primary purpose of the R&D project, or not even be directly foreseeable. Kester (1984) has argued that growth options may account for a large part of project value. BestPharma faced the problem of choosing among several strategic research initiatives. They developed a decision tree representation of the projects, which helped to provide transparency about project value and strategic options. Most importantly, carefully thinking through the tree helped to identify growth options, represented by additional branches in the tree, and to quantify that they represented major sources of value.

Collaborative Prototyping and the Pricing of Custom-Designed Products

A major challenge in the creation of custom-designed products lies in the elicitation of customer needs. As customers are frequently unable to accurately articulate their needs, designers typically create one or several prototypes, which they then present to the customer. This process, which we call collaborative prototyping , allows both parties to anticipate the outcome of the design process. Prototypes have two advantages: They help the customer to evaluate the unknown customized product, and they guide both parties in the search for the ideal product specification. Collaborative prototyping involves two economic agents, with different information structures and different-and potentially conflicting-objective functions. This raises several interesting questions: how many prototypes should be built, who should pay for them, and how should they and the customized product be priced. We show that, depending on the design problem and the market characteristics, the designer should offer prototypes at a profit, at cost, or even for free.

Selecting R&D projects at BMW: a case study of adopting mathematical programming models

Research and development (R&D) project selection is a critical interface between the product development strategy of an organization and the process of managing projects day-to-day. This article describes the project selection problem faced by an R&D group of BMW (Munich, Germany). The problem was structured as minimizing the gap between target performance of the technology to be developed and actual performance of the current technology along chosen criteria. A mathematical programming model helped this organization to increase the transparency of their selection process, which previously had been based on experience coupled with evaluation of individual projects in isolation. Implementation was a success in that the predevelopment group continues to use the model to make better decisions. However, the organization did not use the model for its intended purpose: constrained optimization. The traditional explanation for this partial implementation is that the analytical model did not capture all considerations relevant to optimization (e.g., uncertainty or strategic fit), and that further model refinements are required to achieve further implementation. We offer an alternative explanation, one based on the technology transfer literature.