4344454647484946 of 68
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • apa
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • Other style
More styles
Language
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Other locale
More languages
Output format
  • html
  • text
  • asciidoc
  • rtf
From Reactive to Predictive: Enhancing Sales and Operations Planning in Engineer-To-Order Settings with Machine Learning
Jönköping University, School of Engineering, JTH, Supply Chain and Operations Management.ORCID iD: 0000-0002-4690-5716
2026 (English)Doctoral thesis, comprehensive summary (Other academic)
Abstract [en]

Companies in engineer-to-order (ETO) settings face increasing pressure to enhance effectiveness and efficiency in their supply chain planning (SCP) practices through digital transformation. Yet despite the recognized potential of machine learning (ML), its adoption In SCP practices in ETO settings remains minimal. This implementation gap forms the central motivation fort his thesis.

This thesis explores SCP practices in ETO settings, especially sales and operations planning (S&OP) and request for quotation (RFQ) planning and investigates how ML techniques can be integrated to improve these practices. Through three complementary studies involving nine ETO companies, the research addresses the gap between ML's theoretical potential and practical implementation in complex manufacturing settings. The research employs a multiple case study approach for two of the studies of which one is conducting a mapping of ML techniques in four SCP areas to improve agility and adaptability. The other study is grounded in paradox theory and contingency theory to examine S&OP, RFQ planning and their interaction. Despite mounting pressure for digital transformation, the studies reveal minimal ML adoption across the investigated cases, highlighting a significant implementation contradiction that shapes the research contributions. The third study is focusing on how efficiency and effectivity of SCP in ETO settings enhanced with ML should be measured.

The thesis presents four key findings through five interconnected papers. First, different ETO settings fundamentally impact S&OP and RFQ planning through in total eight hierarchical contingency factors, creating an iterative cycle where practices reshape organizational context and generate new contradictions. Second, performance measurement frameworks require adaptation for ML implementation in ETO settings, with a bidirectional relationship between metrics and technology adoption that challenges traditional linear improvement models. Third, specific ML techniques can address SCP challenges in ETO settings, with systematic mapping across the demand planning, production planning, supply management, and inventory management areas. In addition, specific ML techniques can address contradictions for improved management in S&OP and RFQ planning. Fourth, practical implementation requires a structured actionable framework addressing data integration, targeted applications, scaling, and continuous learning systems. The research makes theoretical contributions by extending the Triple-A framework to ETO settings, advancing paradox theory applications in SCP practices, and challenging established assumptions in performance measurement literature. The integration of paradox and contingency theories provides novel insights into how contradictions and contextual factors dynamically influence SCP practice formation in ETO settings.

Practically, the thesis delivers actionable frameworks, propositions, and implementation guidance for companies pursuing ML adoption in SCP. Successful implementation requires sustained investment in data quality, technical infrastructure, organizational capabilities, and performance maturity, prerequisites that the studied cases had not yet fully established.

Future research should pursue longitudinal studies of actual ML implementations, cross-organizational culture validation, cross-manufacturing setting validation, and quantitative assessment of contradiction management effectiveness.

Abstract [sv]

Företag i engineer-to-order‑miljöer (ETO) står inför en ökande press att förbättra processeffektivitet och förbättra kvalitet och innehåll i sina planeringsprocesser av försörjningskedjan (Supply Chain Planning, SCP) genom digital transformation. Trots den erkända potentialen hos maskininlärning (ML) är användningen av ML inom SCP i ETO‑miljöer fortfarande mycket begränsad. Detta implementeringsgap utgör det centrala skälet till forskningen i denna avhandling.

Avhandlingen undersöker SCP i ETO‑miljöer, med särskilt fokus på Sälj och Verksamhetsplanering (Sales and Operations Planning, S&OP) och planering kring offertförfrågningar (Request for Quotation, RFQ), samt analyserar hur ML‑tekniker kan integreras för att förbättra dessa planeringsprocesser. Genom tre kompletterande studier som omfattar nio ETO‑företag adresserar forskningen gapet mellan ML’s teoretiska potential och dess praktiska tillämpning i komplexa tillverkningsmiljöer. Två av studierna använder fallstudie med flera fall (case) som grund för forskningen, varav den ena genomför en kartläggning av ML‑tekniker inom fyra SCP‑områden i syfte att förbättra förmåga att anpassa sig till kortsiktiga marknadsförändringar (agility) och anpassningsförmåga på lite längre sikt (adaptability). Den andra studien tar sin teoretiska utgångspunkt i paradoxteori och kontingensteori för att analysera S&OP, RFQ‑planering och samspelet mellan dessa. Trots det ökande trycket på digital transformation visar studierna på en mycket begränsad ML‑användning i de undersökta fallen, vilket belyser en motsägelse kring införande av ML som präglar avhandlingens forskningsbidrag. Den tredje studien fokuserar på hur processeffektivitet och processutfall i SCP i ETO‑miljöer, förstärkta med ML, bör mätas.

Avhandlingen presenterar fyra centrala resultat via fem sammanlänkade publikationer. För det första påverkar olika ETO‑miljöer i grunden S&OP‑och RFQ‑planering genom totalt åtta hierarkiska kontingensfaktorer, vilket skapar en iterativ cykel där ett valt processutförande omformar den organisatoriska kontexten och genererar nya motsägelser. För det andra kräver ramverk för prestationsmätning anpassning för ML‑implementering i ETO‑miljöer, med ett dubbelriktat samband mellan mätetal och teknikanvändning som utmanar traditionella linjära förbättringsmodeller. För det tredje kan specifika ML‑tekniker adressera SCP‑utmaningar i ETO‑miljöer, genom en systematisk kartläggning inom områdena efterfrågeplanering, produktionsplanering, försörjningsstyrning och lagerstyrning. Därutöver kan specifika ML‑tekniker hantera motsägelser för förbättrad styrning inom S&OP‑ och RFQ‑planering. För det fjärde kräver praktisk implementering ett strukturerat och handlingsorienterat ramverk som adresserar dataintegration, riktade tillämpningar, skalning och system för kontinuerligt lärande.

Forskningen bidrar teoretiskt genom att vidareutveckla Triple‑A‑ramverket för ETO‑miljöer, fördjupa tillämpningen av paradoxteori inom SCP samt utmana etablerade antaganden inom litteraturen om prestationsmätning. Integrationen av paradox‑ och kontingensteori ger nya insikter i hur motsägelser och kontextuella faktorer dynamiskt påverkar utformningen av SCP i ETO‑miljöer.

Ur ett praktiskt perspektiv levererar avhandlingen ett handlingsorienteratramverk, propositioner och implementeringsvägledning för företag somsträvar efter att införa ML inom SCP. En framgångsrik implementering kräverlångsiktiga investeringar i datakvalitet, teknisk infrastruktur, organisatoriskaförmågor och mognad inom prestationsmätning, förutsättningar som destuderade fallen ännu inte fullt ut hade etablerat.

Framtida forskning bör genomföra longitudinella studier av faktiskaML‑implementationer, validera resultaten över kulturella skillnader mellanorganisationer och tillverkningsmiljöer samt kvantitativt utvärderaeffektiviteten i hantering av motsägelser.

Place, publisher, year, edition, pages
Jönköping: Jönköping University, School of Engineering , 2026. , p. 93
Series
JTH Dissertation Series ; 102
Keywords [en]
Sales and Operations Planning, Supply Chain Planning, Engineer-to-order, Machine Learning, Paradox Theory, Contingency Theory, Triple-A
Keywords [sv]
Sales and Operations Planning, Supply Chain Planning, Engineer‑to‑order, Maskininlärning, Paradoxteori, Kontingensteori, Triple‑A
National Category
Production Engineering, Human Work Science and Ergonomics
Identifiers
URN: urn:nbn:se:hj:diva-73598ISBN: 978-91-89785-37-3 (print)ISBN: 978-91-89785-38-0 (electronic)OAI: oai:DiVA.org:hj-73598DiVA, id: diva2:2093790
Supervisors
Available from: 2026-08-20 Created: 2026-08-20 Last updated: 2026-08-20Bibliographically approved
List of papers
1. How Contradictions Impact Practices in Supply Chain Planning in Engineer-To-Order-Settings: A Multi Case Study
Open this publication in new window or tab >>How Contradictions Impact Practices in Supply Chain Planning in Engineer-To-Order-Settings: A Multi Case Study
2026 (English)In: Advances in Production Management Systems: Cyber-Physical-Human Production Systems: Human-AI Collaboration and Beyond / [ed] H. Mizuyama, E. Morinaga, T. Nonaka, T. Kaihara, G. von Cieminski, D. Romero, Springer, 2026, Vol. 767, p. 424-437Conference paper, Published paper (Other academic)
Abstract [en]

This study examines how inherent contradictions affect supply chain planning (SCP) practices in engineer-to-order (ETO) settings through a multi-case study of five companies. The research identifies sixteen contradictions across four paradox categories: performing, organizing, belonging, and learning paradoxes. The findings reveal that contradictions shape SCP practices while SCP practices simultaneously influence how contradictions are managed. Companies employ four management strategies: acceptance, spatial separation, temporal separation, and synthesis to handle contradictions. One of the approaches acknowledge contradictions as inherent features of ETO settings rather than a problem to eliminate. Companies with structured SCP processes demonstrate better contradiction management capabilities, creating virtuous cycles of continuous improvement. The research contributes to paradox theory by validating the dynamic equilibrium model in SCP contexts and identifying organizational capabilities that enable effective paradox management. For practitioners, the study suggests that explicitly acknowledging contradictions, aligning organizational structures, and developing balanced performance metrics can transform competing demands into sources of competitive advantage rather than operational constraints.

Place, publisher, year, edition, pages
Springer, 2026
Series
IFIP Advances in Information and Communication Technology, ISSN 1868-4238, E-ISSN 1868-422X ; 767
Keywords
Engineer to Order, Paradox theory, Supply Chain Planning, Competition, Information systems, Information use, Case-studies, Continuous improvements, Engineer to orders, Management capabilities, Management strategies, Planning process, Spatial separation, Temporal separation, Engineers
National Category
Business Administration
Identifiers
urn:nbn:se:hj:diva-69798 (URN)10.1007/978-3-032-03542-4_29 (DOI)001583293100029 ()2-s2.0-105015559746 (Scopus ID)978-3-032-03541-7 (ISBN)978-3-032-03542-4 (ISBN)
Conference
44th IFIP WG 5.7 International Conference on Advances in Production Management Systems, Kamakura, 31 August 2025 - 4 September 2025
Funder
Knowledge Foundation
Available from: 2025-09-22 Created: 2025-09-22 Last updated: 2026-08-20Bibliographically approved
2. The planning paradox: Why engineer-to-order companies must embrace contradictions in tactical planning
Open this publication in new window or tab >>The planning paradox: Why engineer-to-order companies must embrace contradictions in tactical planning
2026 (English)In: Production planning & control (Print), ISSN 0953-7287, E-ISSN 1366-5871Article in journal (Refereed) Epub ahead of print
Abstract [en]

Tactical planning practices, like request for quotation (RFQ) planning and sales and operations planning (S&OP) are critical for competitiveness in engineer-to-order (ETO) settings yet remain understudied compared to repetitive manufacturing. This study explores RFQ planning and S&OP practices in ETO settings through integrated paradox and contingency theory lenses, addressing how experienced contradictions affect these practices and how contingency factors influence contradictions and practice design. A multiple case study of five Swedish companies reveals that ETO organizations face nine general and seven context-specific contradictions shaping their tactical planning practices influenced by seven contingency factors grouped into fundamental, mediating and operational levels. The findings challenge static views of organizational fit, revealing a four-stage iterative cycle where contingency factors manifest contradictions that shape RFQ planning and S&OP practices, which subsequently become new contingency factors. This dynamic relationship suggests that effective RFQ planning and S&OP practices in ETO settings requires embracing rather than resolving inherent contradictions through continuous adaptation and sophisticated management. S&OP emerges as a powerful contradiction synthesis mechanism when properly designed for ETO settings. The study extends paradox theory to tactical planning domains, advances contingency theory by revealing practice-contingency co-evaluation.

Place, publisher, year, edition, pages
Taylor & Francis, 2026
Keywords
Request for quotation, engineer-to-order, sales & operations planning, paradox theory, contingency theory
National Category
Production Engineering, Human Work Science and Ergonomics
Identifiers
urn:nbn:se:hj:diva-71131 (URN)10.1080/09537287.2026.2647231 (DOI)001740164200001 ()2-s2.0-105035780721 (Scopus ID)HOA;;71131 (Local ID)HOA;;71131 (Archive number)HOA;;71131 (OAI)
Funder
Knowledge Foundation
Available from: 2026-04-15 Created: 2026-04-15 Last updated: 2026-08-20
3. Measuring efficiency and effectiveness improvements, when implementing machine learning in supply chain planning practices in engineer-to-order settings
Open this publication in new window or tab >>Measuring efficiency and effectiveness improvements, when implementing machine learning in supply chain planning practices in engineer-to-order settings
2025 (English)Conference paper, Oral presentation only (Refereed)
Abstract [en]

Engineer-to-Order (ETO) settings present unique challenges for Supply Chain Planning (SCP) due to their inherent complexity, high customization requirements, and data scarcity. While machine learning (ML) offers promising opportunities to improve planning processes, its implementation in ETO environments remains limited. This paper investigates the relationship between performance measurement and ML adoption in SCP processes within ETO settings. Through a conceptual framework based on literature review and empirical observations from nine industrial cases, the initiation and evaluation of ML implementation in complex planning environments is explored.

Our findings reveal that, contrary to traditional improvement frameworks, ML adoption in ETO settings is not primarily driven by existing performance metrics. Instead, organizations implement ML based on perceived challenges, contradictions in planning processes, and exploratory interest. We identify a bidirectional relationship where performance measurements can drive ML implementation, but also where ML implementation necessitates the establishment of new or refined metrics to evaluate its impact. This relationship is particularly critical for building trust in ML applications, as appropriate performance measurements provide mechanisms for developing both cognitive and emotional trust.

The study contributes a refined conceptual model that integrates perspectives from performance measurement, technology adoption, and trust theory. For practitioners, it offers guidance on initiating ML implementation when performance measurement maturity is low, emphasizing the importance of establishing appropriate metrics even when adoption is driven by challenges rather than measurements. For scholars, this study touches multiple domains and extends current understanding of technology adoption pathways in SCP and highlights the need for ETO-specific approaches to performance measurement when implementing ML solutions in environments characterized by customization, variability, and limited historical data.

Keywords
Performance Measurement, Supply Chain Planning, Engineer-to-order, Machine learning
National Category
Production Engineering, Human Work Science and Ergonomics
Identifiers
urn:nbn:se:hj:diva-70154 (URN)
Conference
15th EDSI Annual Conference, Gothenburg, Sweden, June 1-4, 2025
Funder
Knowledge Foundation
Available from: 2025-11-11 Created: 2025-11-11 Last updated: 2026-08-20Bibliographically approved
4. Managing supply chain planning contradictions in engineer-to-order settings with machine learning
Open this publication in new window or tab >>Managing supply chain planning contradictions in engineer-to-order settings with machine learning
2025 (English)Conference paper, Published paper (Refereed)
Abstract [en]

Purpose

The purpose is to explore how machine learning (ML) can impact or influence the management of Supply Chain Planning (SCP) contradictions in engineer-to-order (ETO) settings.

Design/methodology/approach

A multiple case study approach investigated five Swedish manufacturing companies in ETO-settings but in different industries. Data collection involved semi-structured interviews and questionnaires with SCP-professionals together with archival data, analysed through thematic coding to identify patterns across cases.

Findings

Based on sixteen identified SCP-contradictions across four paradox categories (performing, organizing, belonging and learning), ML-applications were identified that could address each contradiction category, despite limited ML-adoption in the case companies.

Research limitations/implications

The study is limited to Swedish manufacturing companies, potentially affecting generalizability. Future research should empirically test ML-applications' effectiveness in managing SCP-contradictions and explore how it affects the SCP-maturity in organizations in ETO-settings.

Practical implications

Based on the categorization of contradictions the analysis of ML-applications offers targeted guidance and suggestions for ML-methods for managing SCP-contradictions for companies considering ML-adoption.

Original/value

This study extends paradox theory to SCP in ETO-settings and bridges literature on SCP, ETO, and ML by developing a conceptual model illustrating how ML can influence management of SCP-contradiction, potentially transforming contradictions into sources to further develop SCP-practices in ETO-settings.

Keywords
Supply Chain Planning, Engineer-to-order, Machine learning, Paradox Theory
National Category
Production Engineering, Human Work Science and Ergonomics
Identifiers
urn:nbn:se:hj:diva-70157 (URN)
Conference
37th NOFOMA Conference, June 10-12, 2025, Copenhagen, Denmark
Available from: 2025-11-11 Created: 2025-11-11 Last updated: 2026-08-20Bibliographically approved
5. From Reactive to Predictive: Unlocking Machine Learnings Potential in Engineer-to-Order Supply Chains
Open this publication in new window or tab >>From Reactive to Predictive: Unlocking Machine Learnings Potential in Engineer-to-Order Supply Chains
2026 (English)In: Book of Abstracts: 35th Annual IPSERA Conference / [ed] Anni-Kaisa Kähkönen &, International Purchasing and Supply Education and Research Association (IPSERA) , 2026, p. 11-11Conference paper, Oral presentation with published abstract (Refereed)
Abstract [en]

This study explores how machine learning (ML) can enhance supply chain planning (SCP) agility and adaptability in engineer-to-order settings, characterized by uncertainty and complexity. Through multiple-case analysis across heavy machinery manufacturers, three core and four domain specific propositions reveal how ML techniques can theoretically transform four SCP domains. ML driven agility enables rapid responses to disruptions through predictive analytics, while ML driven adaptability supports strategic planning through scenario analysis. Eight contingency factors moderate ML effectiveness, including data quality, organizational culture, and supplier collaboration. An actionable four-phase implementation framework guides organizations in leveraging digital tools to navigate complexity and build resilience.

Place, publisher, year, edition, pages
International Purchasing and Supply Education and Research Association (IPSERA), 2026
Series
IPSERA conference proceedings, E-ISSN 2772-4379
National Category
Software Engineering
Identifiers
urn:nbn:se:hj:diva-73597 (URN)
Conference
35th Annual IPSERA Conference, Conference Theme: "Less is More", 19–22 April 2026, Helsinki–Espoo, Finland
Available from: 2026-08-19 Created: 2026-08-19 Last updated: 2026-08-20Bibliographically approved

Open Access in DiVA

Kappa(3641 kB)20 downloads
File information
File name FULLTEXT01.pdfFile size 3641 kBChecksum SHA-512
1cb6daf0f4b8b212030ea4df67d585f937729edf0bb4c6754d7015ce9e4388f13a8290600735c8c69676ebec5e01e85e3fa2a07d7114f3d01c78945156f95c9c
Type fulltextMimetype application/pdf

Authority records

Ohlson, Nils-Erik

Search in DiVA

By author/editor
Ohlson, Nils-Erik
By organisation
JTH, Supply Chain and Operations Management
Production Engineering, Human Work Science and Ergonomics

Search outside of DiVA

GoogleGoogle Scholar
The number of downloads is the sum of all downloads of full texts. It may include eg previous versions that are now no longer available

isbn
urn-nbn

Altmetric score

isbn
urn-nbn
Total: 1439 hits
4344454647484946 of 68
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • apa
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • Other style
More styles
Language
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Other locale
More languages
Output format
  • html
  • text
  • asciidoc
  • rtf