A complexidade de alguns sistemas biotecnológicos impossibilita seu estudo sem o uso de técnicas de programação matemática avançadas. A quantificação de fluxos metabólicos e a síntese e projeto ótimos de plantas multiproduto são problemas com esta característica, abordados na presente tese. A quantificação de fluxos metabólicos empregando balanços de marcações é representada como um problema de otimização não-linear, o qual se resolve através da minimização da diferença entre as medidas experimentais e as predições do modelo da rede metabólica. Este problema surge da necessidade de se caracterizar o metabolismo mediante a estimação das velocidades das reações bioquímicas. O modelo matemático para problemas deste tipo é composto basicamente por balanços de metabólitos e de isótopos; os primeiros são lineares, enquanto os segundos introduzem não-linearidades ao problema e, neste trabalho, são modelados mediante uma modificação da técnica de matrizes de mapeamento de átomos. Para quantificar os fluxos metabólicos considerando a existência de ótimos locais, desenvolveu-se um algoritmo branch & bound espacial, no qual a busca global é feita mediante a divisão da região de busca (branching) e a geração de seqüências de limites (bounding) que convergem para a solução global. Como estudo de caso, estimaram-se os fluxos no metabolismo central de Saccharomyces cerevisiae. Os resultados confirmam a existência de soluções locais e a necessidade de desenvolver uma estratégia de busca global; a solução global obtida apresenta semelhanças, nos fluxos centrais, com a melhor solução obtida por um algoritmo evolucionário. Quanto aos problemas de síntese e projeto de sistemas biotecnológicos multiproduto, As abordagens mais empregadas para resolve-los são a definição e dimensionamento seqüencial das operações unitárias, e a fixação dos parâmetros de dimensionamento e de estimação do tempo de operação (com valores obtidos em laboratório ou planta piloto); porém ambas abordagens fornecem soluções subótimas. Por outro lado, a solução simultânea da síntese e projeto de sistemas biotecnológicos multiproduto gera modelos misto-inteiros não-lineares (MINLP) de grande porte, devido à combinação das decisões, ligadas à existência de alternativas no processo, com as restrições não-lineares geradas dos modelos das operações. Como estudo de caso considera-se uma planta para produção de insulina, vacina para hepatite B, ativador de plasminogênio tecidual (tissue plasminogen activator) e superóxido dismutase, mediante três hospedeiros diferentes: levedura (S. cerevisiae) com expressão extra ou intracelular, Escherichia coli e células de mamíferos. O projeto deve satisfazer a meta de produção para cada produto, minimizando os custos de capital e selecionando os hospedeiros, as operações e o arranjo dos equipamentos em cada estágio. Os resultados obtidos mostram que a formulação das decisões por abordagem big-M permite resolver o modelo MINLP gerado e que a consideração de múltiplos produtos com seqüências e condições de processamento diferentes gera grande ociosidade nos equipamentos e aumenta o custo total do projeto. Para o estudo de caso observou-se que a alocação de tanques intermediários tem um efeito limitado na diminuição do custo do projeto, porém a implementação simultânea da flexibilização do scheduling, do projeto de equipamentos auxiliares e tanques intermediários permite obter projetos satisfatórios. / The complexity of biotechnological systems does not allow their study without the use of advanced mathematical programming techniques. Metabolic flux quantification and optimal synthesis and design of multiproduct plants are problems with this characteristic, and are addressed in this thesis. The metabolic flux quantification employing labeling balances is formulated as a nonlinear optimization problem that is solved by the minimization of the difference between experimental measurements and predictions of the metabolic network model. This problem is generated by the necessity of estimating the rates of biochemical reactions that characterize the metabolism. The mathematical model for this class of problems is composed by balances of metabolites and isotopes; the former are linear whereas the latter are nonlinear and, in this work, are modeled by a modification of the atom mapping matrix technique. A spatial branch & bound algorithm was developed to quantify the metabolic fluxes, that considers the existence of local optima; in this algorithm, the global search is developed by the division of the searching region (branching) and the generation of sequences of bounds (bounding) that converge to the global solution. As a case study, fluxes in central metabolism of Saccharomyces cerevisiae were estimated. The results confirm the existence of local solutions and the necessity of develop a global search strategy; the central fluxes in the obtained global solution are similar to those ones obtained by an evolutionary algorithm. To solve problems of synthesis and design of multiproduct biotechnological systems, the most employed approaches are the sequential selection and sizing of the unit operations, and the fixing of sizing and time parameters (employing values from laboratory or pilot plants); nevertheless, both approaches generate suboptimal solutions. On the other hand, the simultaneous solution of the synthesis and design of multiproduct biotechnological systems generates large size mixed-integer nonlinear models (MINLP), due to the combination of options into the processing with nonlinear constraints from the operation models. As case study, a plant for production of insulin, hepatitis B vaccine, tissue plasminogen activator and superoxide dismutase was considered, by three hosts: yeast (S. cerevisiae) with extra or intracellular expression, Escherichia coli and mammalian cells. The design must satisfy the production target for each product, minimizing the capital cost and considering the selection of hosts, the operations and the number of parallel units in each stage. The obtained results show that the formulation of decisions by the big-M approach allows the solution of the generated MINLP model and that consideration of several products with different processing sequences and conditions generates large idleness at the equipment and increases the total cost of the design. In the case study it was observed that the allocation of storage tanks has a limited effect on cost reduction, but the simultaneous implementation of flexible scheduling, design of auxiliary equipments and intermediate storage tanks allow the generation of satisfactory designs.

A complexidade de alguns sistemas biotecnológicos impossibilita seu estudo sem o uso de técnicas de programação matemática avançadas. A quantificação de fluxos metabólicos e a síntese e projeto ótimos de plantas multiproduto são problemas com esta característica, abordados na presente tese. A quantificação de fluxos metabólicos empregando balanços de marcações é representada como um problema de otimização não-linear, o qual se resolve através da minimização da diferença entre as medidas experimentais e as predições do modelo da rede metabólica. Este problema surge da necessidade de se caracterizar o metabolismo mediante a estimação das velocidades das reações bioquímicas. O modelo matemático para problemas deste tipo é composto basicamente por balanços de metabólitos e de isótopos; os primeiros são lineares, enquanto os segundos introduzem não-linearidades ao problema e, neste trabalho, são modelados mediante uma modificação da técnica de matrizes de mapeamento de átomos. Para quantificar os fluxos metabólicos considerando a existência de ótimos locais, desenvolveu-se um algoritmo branch & bound espacial, no qual a busca global é feita mediante a divisão da região de busca (branching) e a geração de seqüências de limites (bounding) que convergem para a solução global. Como estudo de caso, estimaram-se os fluxos no metabolismo central de Saccharomyces cerevisiae. Os resultados confirmam a existência de soluções locais e a necessidade de desenvolver uma estratégia de busca global; a solução global obtida apresenta semelhanças, nos fluxos centrais, com a melhor solução obtida por um algoritmo evolucionário. Quanto aos problemas de síntese e projeto de sistemas biotecnológicos multiproduto, As abordagens mais empregadas para resolve-los são a definição e dimensionamento seqüencial das operações unitárias, e a fixação dos parâmetros de dimensionamento e de estimação do tempo de operação (com valores obtidos em laboratório ou planta piloto); porém ambas abordagens fornecem soluções subótimas. Por outro lado, a solução simultânea da síntese e projeto de sistemas biotecnológicos multiproduto gera modelos misto-inteiros não-lineares (MINLP) de grande porte, devido à combinação das decisões, ligadas à existência de alternativas no processo, com as restrições não-lineares geradas dos modelos das operações. Como estudo de caso considera-se uma planta para produção de insulina, vacina para hepatite B, ativador de plasminogênio tecidual (tissue plasminogen activator) e superóxido dismutase, mediante três hospedeiros diferentes: levedura (S. cerevisiae) com expressão extra ou intracelular, Escherichia coli e células de mamíferos. O projeto deve satisfazer a meta de produção para cada produto, minimizando os custos de capital e selecionando os hospedeiros, as operações e o arranjo dos equipamentos em cada estágio. Os resultados obtidos mostram que a formulação das decisões por abordagem big-M permite resolver o modelo MINLP gerado e que a consideração de múltiplos produtos com seqüências e condições de processamento diferentes gera grande ociosidade nos equipamentos e aumenta o custo total do projeto. Para o estudo de caso observou-se que a alocação de tanques intermediários tem um efeito limitado na diminuição do custo do projeto, porém a implementação simultânea da flexibilização do scheduling, do projeto de equipamentos auxiliares e tanques intermediários permite obter projetos satisfatórios. / The complexity of biotechnological systems does not allow their study without the use of advanced mathematical programming techniques. Metabolic flux quantification and optimal synthesis and design of multiproduct plants are problems with this characteristic, and are addressed in this thesis. The metabolic flux quantification employing labeling balances is formulated as a nonlinear optimization problem that is solved by the minimization of the difference between experimental measurements and predictions of the metabolic network model. This problem is generated by the necessity of estimating the rates of biochemical reactions that characterize the metabolism. The mathematical model for this class of problems is composed by balances of metabolites and isotopes; the former are linear whereas the latter are nonlinear and, in this work, are modeled by a modification of the atom mapping matrix technique. A spatial branch & bound algorithm was developed to quantify the metabolic fluxes, that considers the existence of local optima; in this algorithm, the global search is developed by the division of the searching region (branching) and the generation of sequences of bounds (bounding) that converge to the global solution. As a case study, fluxes in central metabolism of Saccharomyces cerevisiae were estimated. The results confirm the existence of local solutions and the necessity of develop a global search strategy; the central fluxes in the obtained global solution are similar to those ones obtained by an evolutionary algorithm. To solve problems of synthesis and design of multiproduct biotechnological systems, the most employed approaches are the sequential selection and sizing of the unit operations, and the fixing of sizing and time parameters (employing values from laboratory or pilot plants); nevertheless, both approaches generate suboptimal solutions. On the other hand, the simultaneous solution of the synthesis and design of multiproduct biotechnological systems generates large size mixed-integer nonlinear models (MINLP), due to the combination of options into the processing with nonlinear constraints from the operation models. As case study, a plant for production of insulin, hepatitis B vaccine, tissue plasminogen activator and superoxide dismutase was considered, by three hosts: yeast (S. cerevisiae) with extra or intracellular expression, Escherichia coli and mammalian cells. The design must satisfy the production target for each product, minimizing the capital cost and considering the selection of hosts, the operations and the number of parallel units in each stage. The obtained results show that the formulation of decisions by the big-M approach allows the solution of the generated MINLP model and that consideration of several products with different processing sequences and conditions generates large idleness at the equipment and increases the total cost of the design. In the case study it was observed that the allocation of storage tanks has a limited effect on cost reduction, but the simultaneous implementation of flexible scheduling, design of auxiliary equipments and intermediate storage tanks allow the generation of satisfactory designs.

本文將設定一 Translog 多產品成本函數模型，以估計台灣旅行業之規模 經濟 (Economies of Scale) 及多樣化經濟 (Economies of Scope)程度 。在 Translog 多產品成本函數模型中，係以廠商所經營之四項業務，包 括來華觀光、出國旅遊、國民旅遊及代辦業務等，當作四種不同的產品， 而要素投入則為勞動與資本租賃。實證結果顯示，台灣旅行業普遍存在規 模經濟的現象。此結果可能導致：第一，所謂``靠行''的形成；第二，規 模較小的廠商可能根據本身的競爭優勢與經營特色而採取市場區隔化策略 ，以服務某特定之顧客群；第三，現有旅行業廠商之間的合併，或是來自 產業外的廠商藉購併進入旅行業市場，以從事多角化經營。另外，就產品 生產效率而言，我們以多樣化經濟存在與否認為，來華觀光與出國觀光之 間、來華觀光與國民旅遊之間，以及出國觀光與國民旅遊之間，因其業務 關聯性不大，而無法發揮資源共用共享的效果，因此不適合聯合生產。而 出國觀光與代辦業務之間，因業務關聯性大，因此若能聯合生產，則必然 能夠產生綜效。至於來華觀光與代辦業務之間，以及國民旅遊與代辦業務 之間，雖無明顯之證據顯示其具有聯合生產之優勢，但若存在閒置產能， 或是因某種業務之專業性不高，致使人力資源之轉換成本不大時，亦不排 除聯合生產之可能。

There are two main views of advertising – the informative view and the persuasive view. This thesis studies aspects of the informative view. One aspect of interest is whether firms can benefit from collusion on advertising even though advertising is only informative. If so, will this enhance or lower welfare? There are several reasons why firms may want to collude on adver­tising. First, the legal field is tilted in favour of nonprice collusion. Second, it is not at all obvious that collusion on price is more profitable than collusion on advertising and third, the analyses of Grossman and Shapiro (1984) and others show that profits can be increased by restricting advertising. In Paper 1, we examine firms’ incentives to collude on advertising and the implications for welfare. We find that collusion on advertising and competition on price is more profitable than competition on both price and advertising. We also find that semicollusion on advertising is detrimental to welfare. This suggests a need for monitoring, especially since it is in the interest of firms to restrict price advertising. We also compare semicollusion on price to semicollusion on advertising. We find that, in general, semicollusion on price does not lead to higher profits compared to semicollusion on advertising. Hence we lend theoretical support to the empirical literature that consistently find evidence of semicollusion on advertising rather than on price. Another important issue concerns the effect, on prices and profits, of ad-vertising only a subset of the product range. Many firms, in particular those in the retail sector, sell a wide variety of products but only advertise a few. Recent empirical evidence suggests that prices of unadvertised products are higher (Milyo and Waldfogel, 1999). Theoretically, little is known. In Paper 2, we study the effects of advertising only a subset of products. We allow for both low and high differentiation and, at the same time, we explicitly model the advertising decision. We find that the extend of differentiation between competing firms plays an important role in the analysis of loss leader pricing. When firms sell products with the same reservation price, loss leader pric­ing obtains only when differentiation is low. When products are less similar however, price competition is less intense and, as a result, firms advertise prices above marginal cost. Our loss leader pricing results enable us to shed some light on the seemingly paradoxical empirical findings in the marketing literature that loss leader pricing fails to increase store traffic, loss leader sales and hence to increase profits. We also consider a different subject – price discrimination. Although it is well understood that movements in the exchange rate have a bearing on firm profitability and hence affect firm behaviour, the role of exchange rate variability in the firm’s choice of the number of varieties to produce has (to my knowledge) not been explored. This, despite the fact that the product mix is an important aspect of firm strategy. By tinkering with the number of varieties, a firm can bolster its ability to extract consumer surplus. In Paper 3, we explore this issue. We show that variability in the exchange rate induces the firm to vertically segment markets (i.e., offer two varieties in each market). This happens because exchange rate variability affects income dispersion and hence the firm’s incentives to extract consumer surplus. To better extract surplus, the firm offers two price-quality menus, high quality variant (priced high) for top-end surplus extraction and a low quality variety (priced low) to address market coverage concerns. / <p>Diss. Stockholm : Handelshögskolan, 2005 S. 3-9: sammanfattning, s. 13-95: 3 uppsatser</p>

Submitted by MÁRCIO GARCIA RODRIGUES null (marciogarcia@maxionsc.com) on 2016-04-28T15:54:31Z No. of bitstreams: 1 Dissertacao_Marcio Garcia_280416.pdf: 2456536 bytes, checksum: b79ef40cb702a054ccb7d7f939704cf6 (MD5) / Approved for entry into archive by Felipe Augusto Arakaki (arakaki@reitoria.unesp.br) on 2016-05-02T12:54:04Z (GMT) No. of bitstreams: 1 rodrigues_mg_me_guara.pdf: 2472389 bytes, checksum: 1c7d0b1b7f5b2d9c139a65d762cda63a (MD5) / Made available in DSpace on 2016-05-02T12:54:04Z (GMT). No. of bitstreams: 1 rodrigues_mg_me_guara.pdf: 2472389 bytes, checksum: 1c7d0b1b7f5b2d9c139a65d762cda63a (MD5) Previous issue date: 2016-04-21 / A crescente variedade de produtos nos sistemas atuais de produção é um grande desafio para a logística interna de abastecimento de materiais numa linha de montagem final, o abastecimento logístico precisa estar em concordância com a programação de produção de forma a disponibilizar para montagem o material necessário, no momento certo e na quantidade correta. A linha de montagem objeto do estudo acompanha esta tendência. Nela são montados modelos variados de chassis. Desta forma, o objetivo geral da pesquisa é desenvolver um modelo de abastecimento para linha de montagem multiprodutos, onde o ponto principal do trabalho é a definição do conceito mais adequado de acordo com o tipo e quantidade do material a ser abastecido. Através de uma pesquisa-ação foi proposto um modelo de abastecimento de linha que demonstrou uma redução de aproximadamente 50% da área ocupada na borda de linha, em consequência da redução de materiais, e aumento de 3% da produtividade, em virtude das reduções das indisponibilidades e perdas por set-up de abastecimento. / The increasing variety of products in current production systems is a big challenge for internal logistics supply of materials to a final assembly line, the logistic supply must be in step with the production program, in order to provide the necessary material for assembly at the right time and in the correct quantities. The assembly line which is the object of this study follows the trend. On it, several models of chassis are assembled. Therefore, the general objective of this research is to develop a supply model for a multi-product assembly line, with the main focus of the work on determining the most suitable concept in terms of the type and amount of material to be supplied. Through research-action, a model to supply the line has been proposed, which demonstrated a reduction of approximately 50% of the area occupied alongside the line, due to the reduction in materials, and an increase of 3% in productivity, owing to the reductions in outages and losses during supply set-up.

For industries with complex processing of raw materials, the problem of cost optimization is particularly relevant, and the mechanism for distributing total costs between jointly produced products still requires more study. For enterprises that process secondary raw materials, the task of allocating costs for each individual type of product is complicated by the lack of direct accounting of costs for each recoverable useful component, the use of all possible relative methods of distribution, and the lack of reasonable criteria for the distribution of complex costs. A model for managing the costs of multi-product production from secondary raw materials is proposed. it is based on the theory of calculating the cost of complex production, including a set of elements, functions, and tools for their implementation, taking into account the flexibility of technologies for extracting useful components from secondary raw materials and allowing to determine the cost intensity of individual types of products. / Для производств с комплексной переработкой сырья проблема оптимизации затрат особенно актуальна, а механизм распределения общих затрат между совместно производимыми продуктами до сих пор требует большего изучения. Для перерабатывающих вторичное сырье предприятий задача распределения затрат на каждый отдельный вид продукции усложняется отсутствием прямого учета затрат по каждому извлекаемому полезному компоненту, использованием всевозможных относительных способов распределения, отсутствие обоснованных критериев распределения комплексных затрат. Предложена модель управления затратами многопродуктового производства из вторичного сырья, базирующаяся на теории калькулирования себестоимости комплексных производств, включающая в себя совокупность элементов, функций, а также инструментарий их реализации, учитывающая гибкость технологий извлечения полезных компонентов из вторичного сырья и позволяющая определить издержкоемкость индивидуальных видов продукции.

In dieser Studie analysieren wir die Interaktionen zwischen Multiproduktwettbewerb und drei Aspekten von Innovation: Research Joint Ventures mit asymmetrischen Unternehmen, Neue Produkteinführung, und Innovation mit Multiproduktunternehmen. Wir untersuchen Mechanismen und Effekte, die einen Einfluß auf die Marktstruktur, das Marktverhalten und das Marktergebnis ausüben, indem wir theoretische Modelle analysieren und empirische strukturelle Modelle schätzen. Diese Arbeit gliedert sich wie folgt: Kapitel 2 verdeutlicht die Interaktion von Multiproduktwettbewerb und der Innovationstätigkeit von Unternehmen. Wir präsentieren die gegenwärtigen theoretischen and empirischen Arbeiten auf dem Gebiet der Innovationen. In Kapitel 3 analysieren wir die Effekte zwischen Multiproduktwettbewerb und Anreize von asymmetrischen Unternehmen, Innovationen zu betreiben und Research Joint Ventures zu gründen. In Kapitel 4 und 5 konzentrieren wir uns auf die Interaktion zwischen Multiproduktwettbewerb und neuer Produkteinführung. Wir präsentieren zwei theoretische Modelle über vertikale Produktdifferenzierung und untersuchen die Anreize für die etablierten Unternehmen neue Produkte mit unterschiedlicher Qualität in den Markt einzuführen. Unternehmen können hierbei entscheiden, ob sie die vorigen Produkte weiterhin im Markt anbieten oder aus dem Markt ziehen. In Kapitel 6 untersuchen wir den Zusammenhang zwischen Multiproduktwettbewerb und Innovation mit Multiproduktunternehmen. Das Verhalten von Multiproduktunternehmen unterscheidet sich von Einzelproduktunternehmen, da Output- und Produkteinführungsentscheidungen auf zentraler Ebene getroffen werden. Wir analysieren die ,Dynamic Random Access Memory' Industrie (DRAM chips sind Halbleiterchips) mit Berücksichtigung von Multiproduktunternehmen und untersuchen Spillovers, Skalenerträge und Lerneffekte, als auch das Verhalten der Unternehmen im Produktmarkt und die Dynamik über den Produktlebenszyklus. In Kapitel 7 fassen wir die Forschungsergebnisse zusammen und bewerten diese im Kontext der gegenwärtigen Forschung. Darüber hinaus schlagen wir weitere Aspekte für zukünftige Forschungsarbeiten auf diesem Gebiet vor. Schließlich beschreiben wir in Kapitel 8 die Datenbanken, die in unseren empirischen Analysen Anwendung finden. email: siebert@medea.wz-berlin.de / In this study we theoretically and empirically analyze the interactions between multiproduct competition and three aspects of innovation: Research Joint Ventures with asymmetric firms, new product introduction, and innovation with multiproduct firms. We investigate the main mechanisms and effects that impact on market structure, behavior and performance by analyzing theoretical and estimating structural models.The thesis is structured as follows: Chapter 2 provides insights into how multiproduct competition may interact with innovation. We survey current theoretical and empirical results on the literature of innovation. In Chapter 3, we analyze the effect of multiproduct competition on the incentives of asymmetric firms to innovate and to form a Research Joint Venture. In Chapters 4 and 5 we concentrate on the interrelation between multiproduct competition and new product introduction. We present two theoretical models of vertical product differentiation and investigate the incentives for incumbent firms to introduce new products in different quality areas. Firms are allowed to keep or withdraw their original products from the market. In Chapter 6 we will focus on the link between multiproduct competition and innovation with multiproduct firms. Multiproduct firms behave differently in the product market compared to single product firms. Decisions for product innovation or output are taken at a centralized level, so that a multiproduct firm takes the effects on other products into account. We analyze the Dynamic Random Access Memory industry (DRAM chips are semiconductor chips) with respect to multiproduct firms and investigate the Spillover, Economies of Scale, and Learning by Doing effects, as well as firms' behaviour in the product market, and the dynamics over the product life cycle. In Chapter 7 we summarize the results, assess the new research findings in the context of contributions to current research, and provide suggestions for future research. Finally, in Chapter 8 we provide a description of the databases we used in our empirical studies. email: siebert@medea.wz-berlin.de

The growth of population and business during the rapid urbanization process in the twentieth century has generated significant demand for transportation. As the demands have grown, road and air transportation are suffering from significant congestion and delays. Continuing expansion of highways and airports has become both expensive and difficult, along with not being able to provide adequate solutions to the growing congestion. One alternative, which is being pursued by many countries, is to invest in efficient high-speed rail networks to meet the pressing demand for mass passenger transportation. This alternative is also one that may have beneficial impacts by reducing energy consumption and alleviating some of the environmental concerns. But to make these infrastructure investments, governments need to make difficult decisions due to the complexity of the industry and technologies involved. This thesis examines decision making by government for such investments. In order to carefully study the industry, we use a two part approach. First, we examine the HSR industry supply-chain. We create a detailed taxonomy of the industry supply-chain and highlight various aspects of the advanced technologies being used, the sophisticated multiproduct nature of the firms, and the diverse international location of the companies. Second, we gather information on all the international HSR contracts between 2001-2011. These contracts enable us to examine business strategies pursued by the major HSR trainset suppliers and component manufacturers, insights into the size of the orders and type of trainsets being delivered, and the formation of partnerships and collaborations to meet the complex demands imposed by Governments when they invite bids for these expensive projects. A detailed examination of the supply-chain shows that the core technologies and competencies are highly concentrated in those countries which historically have had high demand for high-speed rail. Germany, Japan, France, for example, have the highest number of trainset and component suppliers. In more recent years, South Korea and China have emerged as the new frontiers of trainset and components suppliers. This implies that countries who are outside of this group are highly dependent on either importing these technologies and investments or make a concerted effort to develop them via partnerships and technology transfer agreements. Our examination of contracts shows that the size of HSR investment order is important for both business and government strategy. The order size determines the extent of domestic content and production. While many components will inevitably be imported, a larger order size may allow for various components to be manufactured domestically. Order size also appears to influence the nature of partnerships among the firms in the industry. We observe a growing number of HSR investment partnerships among trainset suppliers over time, possibly due to the need to pool risk in these highly complex and uncertain investments, as well as the changing competitive dynamic of HSR markets.