Alan G. Robinson

On the continuous quadratic knapsack problem

We introduce a new algorithm for the continuous bounded quadratic knapsack problem. This algorithm is motivated by the geometry of the problem, is based on the iterative solution of a series of simple projection problems, and is easy to understand and implement. In practice, the method compares favorably to other well-known algorithms (some of which have superior worst-case complexity) on problem sizes up ton = 4000.

The role of front-line ideas in lean performance improvement

Companies are increasingly embracing lean production techniques in their efforts to improve performance. Yet the majority of these companies fail to tap the full potential of “going lean.” By comparing a sample of successful lean initiatives with less successful ones (as defined by the rate of ongoing productivity improvement), this paper identifies a critical component that often is missing in underperforming initiatives—the ability to get large numbers of improvement ideas from front-line employees. High-performing idea systems—which the authors define as those that implement 12 or more ideas per employee per year—were found to be a major factor in successful lean initiatives, for three reasons. First, they created a “lean culture” of daily improvement. Second, they addressed improvement opportunities that were difficult for managers to spot. Third, they promoted rapid organizational learning. In addition to demonstrating the importance of high-performing idea systems for lean, this research provides insight into why such systems are relatively rare: 1) the predominance of the suggestion-box paradigm; and 2) they frequently require significant and difficult changes in operating practices.

Serial and Parallel Equilibration of Large-Scale Constrained Matrix Problems With Application To the Social and Economic Sciences

We have considered the computation of large-scale con strained matrix problems, which arise in numerous ap plications in the social and economic sciences—in par ticular, the general quadratic problem that permits alter native weighting mechanisms on the data and includes the much-studied diagonal problem. The procedure uti lized is row equilibration, column equilibration (RC), which exploits the bipartite network structure of the problem by decomposing it into simpler subproblems that can be solved exactly and in parallel. We compared the efficiency of the RC algorithm to that of a well- known algorithm and established that RC was faster. We used serial RC as a benchmark for our parallel experi mentation and investigated its absolute efficiency on economic data sets and on very large quadratic diagonal problems. We implemented the RC algorithm using Par allel Fortran Prototype on the IBM 3090-600E. The results demonstrate that a constrained matrix problem with as many as a million variables can be solved using RC in minutes of CPU time in a serial environment. Speedups with the parallelized RC algorithm were substantial for the diagonal problems and moderate for the general problems. These computational results broaden the po tential domain of constrained matrix applications.

Managing Continuous Improvement in Vietnam: Unique Challenges and Approaches to Overcome Them

As Vietnam emerges into world markets, Vietnamese organizations are facing the predicament of how to avoid the low-cost labor trap and move up the production value-chain by improving quality and productivity. Continuous improvement (CI) practices have proved fundamental to building and sustaining competitive advantage in other Asian countries such as Japan, Singapore, and Malaysia. If Vietnamese organizations are to pursue higher value-added activities, CI will be critical for them, too. Despite the tremendous interest in Vietnam from the international business community, little research on Vietnamese management practices has been done, particularly on the use of CI techniques. The language barrier, lack of reliable business data, and the culture of government and corporate secrecy in Vietnam make it extremely difficult to do management research in that country. This paper is a report on an ongoing multiyear research project to study the state of CI in Vietnam. It surveys the current management literature relevant to Vietnam. Then, based on fieldwork in five companies and extensive interviews with business leaders and managers, directors of professional and business associations, and other individuals who have extensive knowledge of Vietnam, the authors identify a number of unique challenges to managing CI in that country, and offer approaches to overcome them.

On Ringeisen's isolation game

Abstract We further develop the theory of Ringeisens isolation game on a graph, in which two players alternately ‘switch’ at successive vertices υ not previously switched. The switching operation deletes all edges incident with υ, and creates new edges between υ and those vertices not previously adjacent to it. The game is won when a vertex is first isolated. In this paper we prove the (somewhat surprising) result that with best play such games can be won only either very early on very late, implying that most graphs are nonwinnable by either player.

The set coincidence game: complexity, attainability, and symmetric strategies

Abstract We develop two closely related aspects of the complexity theory of the Set Coincidence Game G(V, W) , in which two players alternately choose elements not previously chosen from a finite, nonempty set V , and W is a given family of nonempty subsets of V (the “inning sets”). The game is won by that player who first adds an element to the set of “chosen” elements S, so that S ϵ W. We first show that the Set Coincidence Game (in an efficient encoding) is complete in PSPACE. We then develop the theory of attainability which we use to give intuition about the structure of minimal forced wins for either player, and about the “difficulty” of the game. We also show that a well-known type of strategy (“Symmetric” or “pairing”) is optimal in the context of our definition of attainability.

Equilibration operators for the solution of constrained matrix problem

The problem of determining a set of coefficients of a matrix which collectively satisfy certain constraints has become known as the constraively Matrix Problem. It has been widely studied because of its frequent appearance as a “core” problem in diverse applications. These include estimation of input-output tables in the regional sciences, of contingency tables in statistics, origin-destination flows in traffic analysis, and social -national accounts in economics.

Continuous improvement in Vietnam: unique approaches for a unique culture

Purpose – This paper aims to enhance the understanding of continuous improvement (CI) practices in Vietnamese organizations. Design/methodology/approach – The study reported here is based on field research using in-depth case studies to investigate the factors underpinning CI effectiveness in Vietnam. Data were collected from direct observations, internal company documents and interviews in six leading Vietnamese companies, as well as interviews with 50 business leaders, managers, practitioners and academics in Vietnam. Findings – This paper identifies the cultural conditions that have most shaped, and continue to shape, the management of CI in Vietnam, and suggests ways that practitioners can design effective CI practices in that country. For example, a very strong top-down management approach seems to be necessary to jump-start CI in Vietnamese organizations. Vietnamese organizations can succeed with CI, but they require substantial investment in human capital to give managers and employees at all level...

On the set coincidence game

Abstract In the Set Coincidence Game G ( V , W ), two players alternately choose elements not previously chosen from a finite, nonempty set V , and W is a given family of nonempty subsets of V (the ‘winning sets’). The winner is that player who first adds an element to the set of ‘chosen’ elements S , so that S ∈ W . This game is closely related to and generalizes Ringeisens Isolation Game on graphs. We develop the theory of G ( V , W ), present and support a conjecture about the structure of minimal forced wins, and then prove a weakened form (the Weak Filter Theorem). It is hoped that the indicated themes about optimal design of forced wins will prove of interest for a variety of combinatorial games.

The isolation game for regular graphs

Abstract In Ringeisens Isolation Game on a graph, two players alternately ‘switch’ at successive vertices v not previously switched: the switching operation deletes all edges incident with v, and creates new edges between v and those vertices not previously adjacent to it. The game is won when a vertex is first isolated. A previous paper established that (with best play) such games can be won only either very early or very late, implying that most graphs are nonwinnable by either player. Here we focus on regular graphs, showing that their Isolation Games cannot be won unless they can be won extremely early, and identifying the winnable regular graphs explicitly.