Instrument suite cost optimisation in a science megaproject

Abstract

This article presents a cost optimisation study of neutron scattering instrumentation at the European Spallation Source (ESS) in Lund, Sweden. This is done by focusing on the main cost drivers for almost half of the instrument hardware-optics and shielding -and trading detailed cost functions against beam transmission functions in a multi-dimensional, yet simple, parameter space. A cost saving of almost 30% is identified. The method is demonstrated in a worked example on a mature instrument design, and there are no reductions in performance. The proposed solution is shown to correspond to a Nash equilibrium of simultaneous shielding and optics strategies, versus the baseline which is trapped by working through sequential strategies. Finally, this cost analysis is bench-marked against, and found to be in agreement with, the costs of operational facilities of a similar class. and optimised simultaneously, around an architectural concept. This is somewhat similar to Brooks' idea of conceptual integrity in software development [4].Achieving this requires co-located personnel, using a common set of tools, and with thorough crosschecking in place, performed by a dedicated team. The suggestion to do this is not taken without a solid basis in existing excellence, since the NIST Center for Neutron Research [5] at NIST (USA) and SINQ [6] at the Paul Scherrer Institut (PSI, Switzerland) both followed this model to some degree.One might expect that a cost reduction carries some sacrifice of scope or performance. This idea is reinforced by many public 'austerity' measures that have taken place since the financial crisis in 2008, but these were often achieved not via efficiency gains but instead from huge budget cuts in critical areas, with proportional reductions in scope. Instead, a simultaneous and architectural shielding and optics strategy will be derived to reduce the costs of these main areas without reducing performance. The other four main cost drivers-choppers, staff, detectors and 'other' that covers admin and overheads, review meetings etc-these are not completely fixed, but they are somewhat less easily optimised architecturally compared to optics and shielding.The savings in the present study come from an expert understanding of the interplay between neutron optics and shielding, and taking a holistic, architectural approach to designing both of those systems in harmony, according to best current practice. As the ESS is an in-kind project, with 15 instruments contributed by foreign partner laboratories, the status-quo approach would be to design each system independently on each project. This would create a duplication of labour, and unnecessary interfaces in the project: both internally between the groups working on optics and shielding, who in some projects are physically located in different countries, and between the project's and its neighbours' shielding activities. Shielding design interfaces in a regulated, nuclear setting are a major risk to any project, let alone a facility attempting to reach new heigh...

Accurate Simulation of Neutrons in Less Than One Minute Pt. 2: Sandman—GPU-Accelerated Adjoint Monte-Carlo Sampled Acceptance Diagrams

“…In a recent paper [13], a cost optimisation of a spallation source facility was given, in which a re-design was performed on a mature instrument design ("the baseline design"). In that costing paper, the goal was to demonstrate the efficacy of the design methodology, and no real discussion was presented on the use of the software itself.…”

“…Using information in the published reports on the instrument in question, it was possible to reconstruct an accurate model of the phase space in good agreement with those the team generated using McStas. Thereafter, the geometry was interactively modified according to the cost optimisation process [13]. The main thrust of the process is to remove the over-engineering in the supermirror m-values and guide dimensions, and curve the guide to block direct line-of-sight view of the source to save money on shielding.…”

“…Nonetheless, using Sandman for only three days, it was possible to double the instrument performance and reduce the cost of the optics and shielding systems by 40%, from e4.4 M to e2.6 M. Three different variants of the beam were examined, with and without benders to curve out of line of sight, along with two different types of expansion/compression geometry. The best option was then manually refined according to the principles described in the previous work [13]. If one examines the resulting phase space in Figure 3, it is interesting that the brightest parts of the phase space increased only by a small amount.…”

Productivity Curve and Social Network Analysis in Science Megaproject Management

“…This is the third article in a series examining scientific megaprojects. The first paper reported the development of a technical strategy [1] for scientific megaproject instrumentation, where each unit is a bespoke design as part of a larger suite. The second [2] demonstrated the use of metaheuristics in the optimal design of the facility, where the performance metrics are non-linear functions with many parameters.…”

“…Decision making processes are hampered by many stakeholders possessing conflicting interests. Whilst it is to be expected that reaching new scientific horizons requires some novelty and non-standard technology, there is also widespread uniqueness bias -in the case of instrumentation sub-projects, for example, uniqueness bias may result in some fraction of project failing to consider standard solutions correctly in a game tree [1] and instead pursue bespoke solutions at considerably greater cost for little gain, or even strive towards a demonstrably inferior result [5].…”