Quality assurance of qualitative analysis in the framework of the European project ’MEQUALAN’

 The European Commission has supported the G6MA-CT-2000–01012 project on ”Metrology of Qualitative Chemical Analysis” (MEQUALAN), which was developed during 2000–2002. The final result is a document produced by a group of scientists with expertise in different areas of chemical analysis, metrology and quality assurance. One important part of this document deals, therefore, with aspects involved in analytical quality assurance of qualitative analysis. This article shows the main conclusions reported in the document referring to the implementation of quality principles in qualitative analysis: traceability, reliability (uncertainty), validation, and internal/external quality control for qualitative methods. Received: 15 October 2002 Accepted: 20 October 2002 This paper is a summary of the Quality Assurance section included in the final report of the MEQUALAN project. The authors of this paper correspond to the members of the MEQUALAN Consortium. One of them (K.H.) does not fully agree with some parts of the text. Correspondence to A. Ríos     

A genetic algorithm for the resource constrained multi-project scheduling problem

This paper presents a genetic algorithm for the resource constrained multi-project scheduling problem. The chromosome representation of the problem is based on random keys. The schedules are constructed using a heuristic that builds parameterized active schedules based on priorities, delay times, and release dates defined by the genetic algorithm. The approach is tested on a set of randomly generated problems. The computational results validate the effectiveness of the proposed algorithm.     

R&D pipeline management: Task interdependencies and risk management

Maintaining a rich research and development (R&D) pipeline is the key to remaining competitive in many industrial sectors. Due to its nature, R&D activities are subject to multiple sources of uncertainty, the modeling of which is compounded by the ability of the decision maker to alter the underlying process. In this paper, we present a multi-stage stochastic programming framework for R&D pipeline management, which demonstrates how essential considerations can be modeled in an efficient manner including: (i) the selection and scheduling of R&D tasks with general precedence constraints under pass/fail uncertainty, and (ii) resource planning decisions (expansion/contraction and outsourcing) for multiple resource types. Furthermore, we study interdependencies between tasks in terms of probability of success, resource usage and market impact. Finally, we explore risk management approaches, including novel formulations for value at risk and conditional value at risk.     

Modeling methods and a branch and cut algorithm for pharmaceutical clinical trial planning using stochastic programming

We discuss methods for the solution of a multi-stage stochastic programming formulation for the resource-constrained scheduling of clinical trials in the pharmaceutical research and development pipeline. First, we present a number of theoretical properties to reduce the size and improve the tightness of the formulation, focusing primarily on non-anticipativity constraints. Second, we develop a novel branch and cut algorithm where necessary non-anticipativity constraints that are unlikely to be active are removed from the initial formulation and only added if they are violated within the search tree. We improve the performance of our algorithm by combining different node selection strategies and exploring different approaches to constraint violation checking.     

Selection among ranked projects under segmentation,policy and logical constraints

The paper proposes a method for project selection under a specific decision situation, where a final selection is guided by two aspects: (1) satisfaction of certain segmentation, policy and/or logical constraints, and (2) assurance that the individual evaluation of the projects is respected to the maximum degree. This approach is somewhat different than the usual portfolio optimization, where combinations of projects are compared without special concern on respecting the project’s ranking. The entire process is implemented in two phases: the projects are first ranked, usually through a multicriteria approach. The obtained complete preorder of the projects is then used in an integer programming module in order to effectively drive the final selection that satisfies the segmentation and/or logical constraints. The innovative part of the proposed approach is the way it overcomes the well-known bias towards low cost projects which is caused by the knapsack formulation commonly used in the integer programming phase. Actually this is the main source of divergence between the final selection and the initial complete preorder of the projects. The proposed method improves an agreement between the final selection of projects obtained from the integer programming model and the ranking obtained from the multicriteria approach.     

Scheduling Markovian PERT networks to maximize the net present value

We examine project scheduling with net present value objective and exponential activity durations, using a continuous-time Markov decision chain. On the basis of a judicious partitioning of the state space, we achieve a significant performance improvement as compared to the existing algorithms.     

Lower bound for the mean project completion time in dynamic PERT networks

We apply the stochastic dynamic programming to obtain a lower bound for the mean project completion time in a PERT network, where the activity durations are exponentially distributed random variables. Moreover, these random variables are non-static in that the distributions themselves vary according to some randomness in society like strike or inflation. This social randomness is modelled as a function of a separate continuous-time Markov process over the time horizon. The results are verified by simulation.     

An automatic algorithm selection approach for the multi-mode resource-constrained project scheduling problem

This paper investigates the construction of an automatic algorithm selection tool for the multi-mode resource-constrained project scheduling problem (MRCPSP). The research described relies on the notion of empirical hardness models. These models map problem instance features onto the performance of an algorithm. Using such models, the performance of a set of algorithms can be predicted. Based on these predictions, one can automatically select the algorithm that is expected to perform best given the available computing resources. The idea is to combine different algorithms in a super-algorithm that performs better than any of the components individually. We apply this strategy to the classic problem of project scheduling with multiple execution modes. We show that we can indeed significantly improve on the performance of state-of-the-art algorithms when evaluated on a set of unseen instances. This becomes important when lots of instances have to be solved consecutively. Many state-of-the-art algorithms perform very well on a majority of benchmark instances, while performing worse on a smaller set of instances. The performance of one algorithm can be very different on a set of instances while another algorithm sees no difference in performance at all. Knowing in advance, without using scarce computational resources, which algorithm to run on a certain problem instance, can significantly improve the total overall performance.     

Complexity results for the linear time–cost tradeoff problem with multiple milestones and completely ordered jobs

We consider two linear project time–cost tradeoff problems with multiple milestones. Unless a milestone is completed on time, penalty costs for tardiness may be imposed. However, these penalty costs can be avoided by compressing the processing times of certain jobs that require additional resources or costs. Our model describes these penalty costs as the total weighted number of tardy milestone. The first problem tries to minimize the total weighted number of tardy milestones within the budget for total compression costs, while the second problem tries to minimize the total weighted number of tardy milestones plus total compression costs. We develop a linear programming formulation for the case with a fixed number of milestones. For the case with an arbitrary number of milestones, we show that under completely ordered jobs, the first problem is NP-hard in the ordinary sense while the second problem is polynomially solvable.     

Scenario-based portfolio selection of investment projects with incomplete probability and utility information

In the selection of investment projects, it is important to account for exogenous uncertainties (such as macroeconomic developments) which may impact the performance of projects. These uncertainties can be addressed by examining how the projects perform across several scenarios; but it may be difficult to assign well-founded probabilities to such scenarios, or to characterize the decision makers’ risk preferences through a uniquely defined utility function. Motivated by these considerations, we develop a portfolio selection framework which (i) uses set inclusion to capture incomplete information about scenario probabilities and utility functions, (ii) identifies all the non-dominated project portfolios in view of this information, and (iii) offers decision support for rejection and selection of projects. The proposed framework enables interactive decision support processes where the implications of additional probability and utility information or further risk constraints are shown in terms of corresponding decision recommendations.