Designing a Global Manufacturing Network for Local Final Assembly
Bibliographic record
Abstract
This master’s thesis explores how a Global Manufacturing Network can be designed with predefined requirements for local final assembly at an overseas plant. This was done in collaboration with Alfa Laval, where the global manufacturing network of Freshwater Generators (FWGs) was analyzed. The study evaluated options of sourcing directly from suppliers and via the main factory to different extents. A competitive solution is the adoption of Knocked-Down, a methodology to ship all the required parts in arranged kits from the global manufacturing site to the local, overseas sites in the network. These types of networks imply higher throughput times caused by added processes, such as kitting, packaging, and sorting. Therefore, these processes should be allocated appropriately, capitalizing on differences in salary levels and expertise between the sites in the network. A literature review was conducted to establish solid knowledge within the areas of Global Manufacturing Networks, Global Trade Rules, and Knocked-down Supply Chains. Thereafter, an understanding of the global manufacturing network for FWGs was established, through triangulation of qualitative and quantitative data. Interviews with 18 key employees and managers at Alfa Laval were conducted together with an extensive archival research. A systematic analytical framework was followed, and a variety of different network designs was compared and evaluated against each other. This was done by applying existing, theoretical frameworks and conducting a cost driver analysis. For the specific case, the cost drivers were quantified, and an analysis was conducted to clarify how they can be reduced, through improved operations efficiency. The study suggests that Alfa Laval should strive to design a Hub & Spoke network, utilizing its main global site to achieve high economies of scale, while still having local presence. Aligned with theoretical frameworks, adopting a Semi Knocked-Down (SKD) setup was the most cost-efficient solution that fulfills the predefined requirements. The network type includes pre-assembly at the main site, which allows considerable synergies in assembly, sourcing, and inventory costs, together with a remarkable reduction of the kitting time. For the specific case, total cost per unit could be reduced by 10-15% by switching to SKD from the current Completely Knocked-Down (CKD). The total cost per unit could be reduced further by a list of recommended actions, eliminating double work, reducing inefficiencies, and making well-informed trade-offs. The existing theory on knocked-down supply chains has mainly analyzed the automotive industry and how companies can adopt these network designs to enter emerging markets. This master’s thesis of a non-automotive company shows that knocked-down can be applied for already established markets. It can be used to improve competitiveness through increased utilization of economies of scale and synergies between the sites in the global network. The study shows that Customs Duties and Value-added Taxes can have a big impact on the total manufacturing cost and that global manufacturers should pursue “Tax-policies engineering”. Lastly, this study shows that the choice of a knocked-down network design can be influenced by the Rules of Origin. Keywords: Global Manufacturing Network, Knocked-down Supply Chains, Semi-Knocked Down, Completely Knocked-Down, Customs Duties, Value-added Taxes (VAT), Rules of Origin
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.004 | 0.004 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.003 | 0.001 |
| Scholarly communication | 0.004 | 0.005 |
| Open science | 0.001 | 0.004 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.016 | 0.003 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".