Selection of Warehouse Automation System(s) for component storage - in a multiple assembly line manufacturing context Master’s thesis in Supply Chain Management ERLAND NILSSON VIKRAMAN JAYARAMAN DEPARTMENT OF TECHNOLOGY MANAGEMENT AND ECONOMICS DIVISION OF SUPPLY AND OPERATIONS MANAGEMENT CHALMERS UNIVERSITY OF TECHNOLOGY Gothenburg, Sweden 2021 www.chalmers.se Report No. E2021:045 www.chalmers.se Master’s thesis 2021 Report No. E2021:045 Selection of Warehouse Automation System(s) for component storage -in a multiple assembly line manufacturing context Erland Nilsson Vikraman Jayaraman Department of Technology Management and Economics Division of Supply and Operations Management Chalmers University of Technology Gothenburg, Sweden 2021 Selection of Warehouse Automation System(s) for component storage -in a multiple assembly line manufacturing context Erland Nilsson Vikraman Jayaraman © Erland Nilsson & Vikraman Jayaraman, 2021. Supervisor: Lars Svedung, Parker Hannifin Manufacturing Sweden AB (Pump & Motor Division Europe) Supervisor & Examiner: Robin Hanson, Department of Technology Management and Economics, Chalmers Master’s Thesis 2021 Report No. E2021:045 Department of Technology Management and Economics Division of Supply and Operations Management Chalmers University of Technology SE-412 96 Gothenburg Telephone +46 31 772 1000 Cover: The theoretical framework for selection of Warehouse automation system(s) for component storage developed in this master’s thesis. Printed by Chalmers Reproservice Gothenburg, Sweden 2021 iv Selection of Warehouse Automation System(s) for component storage -in a multiple assembly line manufacturing context Authors: Erland Nilsson & Vikraman Jayaraman Department of Technology Management and Economics Chalmers University of Technology Abstract The purpose of this master’s thesis is to explore how the case company, which is a large scale manufacturer, can improve its internal materials supply operation by increasing the level of automation of the component storage systems. This master’s thesis contain com- prehensive and cohesive recommendations for what performance requirements that are imposed on warehouse automation system(s), which will be used for component stor- age by a case company within a multiple assembly line manufacturing context. Such performance requirements are imposed by the Operations / Factory Environment, such as: the Production layout; Manufacturing strategies applied; Characteristics of demand (i.e., the Four Vs: Volume; Variation; Variability; and Visibility); as well as the Char- acteristics of the assembly lines that the stored components are designated for. Further requirements for what warehouse automation system(s) to select for component storage purposes, are imposed by the Material Supply System, like: the Material feeding principles applied; Handling equipment; Storage systems; Packaging and unit loads; and Replenish- ment methods utilized. In addition, there are requirements imposed by the Performance objectives (i.e., Quality, Speed, Dependability, Flexibility and Ergonomics) for the inter- nal material supply processes, which all should be considered when selecting warehouse automation systems for component storage. A theoretical framework has been developed based on all these requirements imposed, which can be used as a point of departure when selecting warehouse automation system for component storage for any company in a sim- ilar manufacturing context. This master’s thesis, also include recommendations for what type of warehouse automation system(s) the case company should select for component storage, based on what type of system that best fulfills the performance requirements imposed. Further, market research has been conducted to identify the best suited ware- house automation system(s) based on all the requirements imposed. In this case, the recommendations are that the case company selects the vertical carousel for component storage within the case facility. Further, it is recommended that each assembly line has a dedicated vertical carousel each. These recommendations provides the best solution, considering the performance requirements imposed on the warehouse automation systems for component storage. In addition, the expected consequences for the company’s internal material supply operation related to implementing these recommendations are described as well. Keywords: Warehouse automation, Component storage automation, Automated compo- nent storage, Automated storage and retrieval systems, AS/RS, Manufacturing logistics, Internal logistics, Material supply operation, Material feeding processes. v Acknowledgements This master’s thesis project was initiated by Parker Hannifin Manufacturing Sweden AB (Pump & Motor Division Europe). Further, this research project has been conducted at Chalmers University of Technology, within the Department of Technology Management and Economics, as a final part of the master program in Supply Chain Management. We would especially like to thank our supervisor at the case company, Lars Svedung at Parker Hannifin Manufacturing Sweden AB (Pump & Motor Division Europe), who pro- vided us the opportunity to conduct this research project and who have supported us with valuable information and feedback during the project proceeding. Also, we would like to thank Lars Markström and all other members of Parker Hannifin who have supported us in data collection, by agreeing to interviews and site visits, and by providing other necessary information throughout the project. We would also like to thank our supervisor and examiner, Robin Hanson at the Division of Supply and Operations Management at Chalmers University of Technology, for productive and insightful feedback throughout this thesis project. Erland Nilsson & Vikraman Jayaraman, Gothenburg, June 2021 vii Contents List of Figures xv List of Tables xvii 1 Introduction 1 1.1 Background . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 1.1.1 The case company’s situation . . . . . . . . . . . . . . . . . . . . . . 3 1.1.2 The case company’s situation in relation to the wider context . . . . 5 1.2 Aim . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5 1.3 Research questions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6 1.4 Limitations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6 1.5 Sustainability - Ethical, societal and ecological aspects . . . . . . . . . . . . 7 2 Methodology 9 2.1 Project procedure . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9 2.2 Empirical data collection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11 3 Theoretical Framework 13 3.1 Operations / Factory Environment . . . . . . . . . . . . . . . . . . . . . . . 13 3.1.1 The Customer order decoupling point & Manufacturing strategies . 13 3.1.2 Characteristics of demand - influences material supply processes . . 16 3.1.3 Performance Objectives - fit production system design with manu- facturing objectives . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16 3.1.3.1 Five generic performance objectives . . . . . . . . . . . . . 16 3.1.3.2 Ergonomics - the importance of human well-being for high performing material handling . . . . . . . . . . . . . . . . . 19 3.1.4 Performance measurement systems - monitoring performance within manufacturing logistics . . . . . . . . . . . . . . . . . . . . . . . . . 21 3.1.4.1 Criteria for a performance measurement system . . . . . . 21 3.1.4.2 Three types of logistics performance variables . . . . . . . . 22 3.1.4.3 Key performance indicators within manufacturing logistics 23 3.1.4.4 Conflicting and complementary relationships between per- formance measures . . . . . . . . . . . . . . . . . . . . . . . 25 3.1.5 Process Improvements - reaching performance objectives . . . . . . . 26 3.1.5.1 Designing a production system - three ways to handle pro- cess variability . . . . . . . . . . . . . . . . . . . . . . . . . 26 3.1.5.2 Process variability - an underlying driver of manufacturing performance . . . . . . . . . . . . . . . . . . . . . . . . . . 27 3.1.5.3 Four main sources of process variability . . . . . . . . . . . 28 ix Contents 3.1.6 Lean logistics - reducing non value-adding wastes within internal logistics processes . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28 3.1.6.1 Waste - activities that add cost but no value . . . . . . . . 29 3.1.6.2 Waste elimination - three strategic focus areas . . . . . . . 30 3.1.6.3 Lean Six Sigma logistics - a combination of two process improvement methods . . . . . . . . . . . . . . . . . . . . . 31 3.1.7 Assembly lines - the internal customers of stored components . . . . 31 3.1.7.1 Single-model, mixed-model, and multi-model assembly lines 32 3.1.7.2 Line balancing - smoothing workloads between process stages 32 3.1.7.3 Takt time and level production - matching production pro- cesses with customer demand . . . . . . . . . . . . . . . . . 33 3.1.7.4 Paced line vs. Unpaced asynchronous line vs. Unpaced synchronous line . . . . . . . . . . . . . . . . . . . . . . . . 34 3.2 Material Supply System . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35 3.2.1 Material feeding - how components reach the assembly line . . . . . 35 3.2.1.1 Continuous Supply - a material feeding principle . . . . . . 36 3.2.1.2 Batch Supply - a material feeding principle . . . . . . . . . 36 3.2.1.3 Kitting Supply - a material feeding principle . . . . . . . . 36 3.2.1.4 Sequencing - to coordinate material supply with the se- quence of material demand . . . . . . . . . . . . . . . . . . 37 3.2.1.5 Hybrid feeding policy . . . . . . . . . . . . . . . . . . . . . 37 3.2.2 Material handling and Transportation . . . . . . . . . . . . . . . . . 37 3.2.3 Storage and inventory . . . . . . . . . . . . . . . . . . . . . . . . . . 39 3.2.3.1 Centralized versus Decentralized storage - benefits and dis- advantages . . . . . . . . . . . . . . . . . . . . . . . . . . . 39 3.2.3.2 Grouping strategies inside a warehouse . . . . . . . . . . . 39 3.2.3.3 High space utilization versus easy material handling - a storage design conflict . . . . . . . . . . . . . . . . . . . . . 39 3.2.3.4 Warehouse main activities - from receival to shipping . . . 40 3.2.3.5 Storage Assignment methods - determining the location of items stored . . . . . . . . . . . . . . . . . . . . . . . . . . 40 3.2.4 Packaging - implications on a material supply operation . . . . . . . 42 3.2.5 Manufacturing Planning and Control . . . . . . . . . . . . . . . . . . 43 3.3 Automation of internal logistics . . . . . . . . . . . . . . . . . . . . . . . . . 44 3.3.1 Automated Equipment . . . . . . . . . . . . . . . . . . . . . . . . . . 44 3.3.1.1 Automation with Robots . . . . . . . . . . . . . . . . . . . 44 3.3.1.2 Automation in transportation . . . . . . . . . . . . . . . . 44 3.3.1.3 Automation of order picking system . . . . . . . . . . . . . 45 3.3.1.4 Selection and design of an order picking system . . . . . . 47 3.3.2 Automation within data collection technologies . . . . . . . . . . . . 48 3.3.2.1 Labelling technologies . . . . . . . . . . . . . . . . . . . . . 48 3.3.2.2 Picking technologies . . . . . . . . . . . . . . . . . . . . . . 48 3.3.3 Automation in Information, Communication and Technology (ICT) . 49 3.4 Application of the Theoretical Framework . . . . . . . . . . . . . . . . . . . 50 4 Empirical data 51 4.1 The case company’s background - a large scale manufacturer . . . . . . . . 51 4.1.1 Four main product groups - hydraulic pumps and motors . . . . . . 51 4.2 The existing production layout - an overview of the case facility . . . . . . . 52 x Contents 4.2.1 A potential future state production layout . . . . . . . . . . . . . . . 53 4.3 Goods receiving . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 53 4.4 The rack system and load carriers used . . . . . . . . . . . . . . . . . . . . . 54 4.4.1 The E-serie load carriers . . . . . . . . . . . . . . . . . . . . . . . . . 55 4.4.2 The F-serie load carriers . . . . . . . . . . . . . . . . . . . . . . . . . 56 4.4.3 The P-serie load carriers . . . . . . . . . . . . . . . . . . . . . . . . . 56 4.5 Storage areas . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 56 4.5.1 Component inventory build up before for summer . . . . . . . . . . 58 4.5.2 Storage replenishment methods . . . . . . . . . . . . . . . . . . . . . 58 4.5.2.1 Kanban cards - used for triggering production of in-house manufactured components . . . . . . . . . . . . . . . . . . 58 4.5.2.2 Hybrid Kanban cards - used for triggering call-offs of pur- chased components . . . . . . . . . . . . . . . . . . . . . . 59 4.5.2.3 Material cards - used for signaling the need for component replenishment . . . . . . . . . . . . . . . . . . . . . . . . . 60 4.5.2.4 Lot numbers - a system used for replenishing individual unit loads of components . . . . . . . . . . . . . . . . . . . 60 4.5.2.5 Stock balance replenishment - based on the stored volume of unique item numbers . . . . . . . . . . . . . . . . . . . . 61 4.5.2.6 Two-bin system - used for some smaller materials . . . . . 61 4.6 The assembly lines - the internal customers of components stored . . . . . . 61 4.6.1 Product group AAA - the AAA:1, AAA:2, and AAA:3 assembly line 62 4.6.2 Product group BBB - the BBB:1 and BBB:2 assembly line . . . . . 66 4.6.3 Product group CCC - the CCC:1 assembly line . . . . . . . . . . . . 69 4.6.4 Product group DDD - the DDD:1 assembly line . . . . . . . . . . . . 71 4.7 Equipments/Automation used . . . . . . . . . . . . . . . . . . . . . . . . . . 73 4.8 Logistics performance measurements applied . . . . . . . . . . . . . . . . . . 74 4.8.1 Revenue influencing logistics variables . . . . . . . . . . . . . . . . . 74 4.8.2 Cost influencing logistics variables . . . . . . . . . . . . . . . . . . . 75 4.8.3 Asset influencing variables . . . . . . . . . . . . . . . . . . . . . . . . 75 4.8.4 Ergonomic influencing variables . . . . . . . . . . . . . . . . . . . . . 76 5 Market Research 77 5.1 Mini-Load AS/RS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 78 5.1.1 Dematic Rapidstore . . . . . . . . . . . . . . . . . . . . . . . . . . . 79 5.1.2 Swisslog Tornado . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 79 5.2 Shuttle- and Robot- based AS/RS . . . . . . . . . . . . . . . . . . . . . . . 80 5.2.1 Swisslog Cyclone . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 80 5.2.2 SSI Flexi . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 81 5.3 Carousel-based AS/RS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 81 5.3.1 Horizontal Carousel -Kardex Remstar . . . . . . . . . . . . . . . . . 82 5.3.2 Vertical Carousel -Megamat RS . . . . . . . . . . . . . . . . . . . . . 82 5.4 Vertical Lift Module (VLM) AS/RS . . . . . . . . . . . . . . . . . . . . . . 83 5.5 Autostore AGV . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 84 6 Analysis 85 6.1 Part 1 - Identification and analysis of requirements for the selection of warehouse automation system(s) for component storage . . . . . . . . . . . 85 6.1.1 Operations / Factory Environment . . . . . . . . . . . . . . . . . . . 86 xi Contents 6.1.1.1 Implications of the existing Production layout - for the selection of warehouse automation system(s) . . . . . . . . 86 6.1.1.2 Implications of the existing Manufacturing strategies ap- plied - for the selection of warehouse automation system(s) 101 6.1.1.3 Implications of the existing Characteristics of demand (i.e., the four Vs) - for the selection of warehouse automation system(s) . . . . . . . . . . . . . . . . . . . . . . . . . . . . 102 6.1.1.4 Implications of the Characteristics of assembly lines - for the selection of warehouse automation system(s) . . . . . . 112 6.1.2 Material Supply System . . . . . . . . . . . . . . . . . . . . . . . . . 116 6.1.2.1 Material feeding principles . . . . . . . . . . . . . . . . . . 116 6.1.2.2 Handling equipment . . . . . . . . . . . . . . . . . . . . . . 117 6.1.2.3 Storage systems . . . . . . . . . . . . . . . . . . . . . . . . 118 6.1.2.4 Packaging and unit loads . . . . . . . . . . . . . . . . . . . 118 6.1.2.5 Replenishment methods . . . . . . . . . . . . . . . . . . . . 119 6.1.3 Performance Objectives . . . . . . . . . . . . . . . . . . . . . . . . . 120 6.1.3.1 Quality . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 120 6.1.3.2 Speed . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 120 6.1.3.3 Dependability . . . . . . . . . . . . . . . . . . . . . . . . . 121 6.1.3.4 Flexibility . . . . . . . . . . . . . . . . . . . . . . . . . . . 122 6.1.3.5 Ergonomics . . . . . . . . . . . . . . . . . . . . . . . . . . . 122 6.2 Part 2 - Selection of warehouse automation system(s) for component storage 123 6.2.1 Performance measurements with mediating effects on the case com- pany’s main logistics KPIs . . . . . . . . . . . . . . . . . . . . . . . . 123 6.2.2 Inter-dependencies between different internal material supply deci- sions - at the case facility . . . . . . . . . . . . . . . . . . . . . . . . 124 6.3 The main recommendations - for a new internal material supply operation . 126 6.3.1 Centralized storage policy and kitting supply for the AAA:1, AAA:2, AAA:3, CCC:1, and DDD:1 assembly line . . . . . . . . . . . . . . . 127 6.3.2 Decentralized storage policy and continuous supply for the BBB:1 and BBB:2 assembly line . . . . . . . . . . . . . . . . . . . . . . . . 130 6.3.3 The use of vertical carousels for all component picking/kitting pro- cesses . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 132 6.3.4 Strengths and weaknesses of the main recommendations . . . . . . . 133 6.3.5 Possible but not recommended alternatives . . . . . . . . . . . . . . 139 6.3.5.1 Rejected general storage policies . . . . . . . . . . . . . . . 139 6.3.5.2 Rejected material feeding principles . . . . . . . . . . . . . 140 6.3.5.3 Rejected warehouse automation system alternatives . . . . 140 7 Conclusion 143 8 Discussion 147 References 151 A Appendix - Annual production data (AAA:1 assembly line) I B Appendix - Variation in volume assembled (AAA:1 assembly line) III C Appendix - Annual production data (AAA:2 assembly line) V xii Contents D Appendix - Variation in volume assembled (AAA:2 assembly line) VII E Appendix - Annual production data (AAA:3 assembly line) IX F Appendix - Variation in volume assembled (AAA:3 assembly line) XI G Appendix - Annual production data (BBB:1 assembly line) XIII H Appendix - Variation in volume assembled (BBB:1 assembly line) XV I Appendix - Annual production data (BBB:2 assembly line) XVII J Appendix - Variation in volume assembled (BBB:2 assembly line) XIX K Appendix - Annual production data (CCC:1 assembly line) XXI L Appendix - Variation in volume assembled (CCC:1 assembly line) XXIII M Appendix - Annual production data (DDD:1 assembly line) XXV N Appendix - Variation in volume assembled (DDD:1 assembly line) XXVII O Appendix - Numerical Calculations (1 of 3) XXIX P Appendix - Numerical Calculations (2 of 3) XXX Q Appendix - Numerical Calculations (3 of 3) XXXII R Appendix - Unit Load Calculation XXXIII xiii Contents xiv List of Figures 3.1 The Customer Order Decoupling Point (CODP) in relation to production and de- livery lead time. Adapted from Wikner and Rudberg (2005). . . . . . . . . . . . . 14 3.2 The Customer Order Decoupling Point (CODP) in relation to four manufacturing strategies. Adapted from Wikner and Rudberg (2005) . . . . . . . . . . . . . . . 15 3.3 Describes the relationship between Throughput efficiency, Capacity utilization and Lead time. Adapted from New (1993). . . . . . . . . . . . . . . . . . . . . . . . 25 3.4 The relationship between Queue capacity and average work-in-process, average time in system, average utilization, and average throughput, respectively. Adapted from Crandell & Burwell (1993). . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26 3.5 Show the relationship between process variability and manufacturing plant perfor- mance. Adapted from Mapes et al. (2000). . . . . . . . . . . . . . . . . . . . . . 28 3.6 Example of an unbalanced vs. balanced assembly line. Adapted from Baudin (2002) 33 3.7 Example of a cycle time mix that is unleveled vs. leveled to Takt time. Adapted from Linc & Cochran (2000). . . . . . . . . . . . . . . . . . . . . . . . . . . . 34 3.8 Example of a Product mix that is unleveled vs. leveled to Customer demand rate. Adapted from Linck & Cochran (1999). . . . . . . . . . . . . . . . . . . . . . . 34 3.9 Categorization of material feeding principles. (Source: Johansson, 1991). . . . . . 35 3.10 Selection of equipment based on use (Source: Hales and Anderson, 2001). . . . . . 38 3.11 Classification of order picking system (OPS). (Source: Dallari et al., 2009) . . . . 46 3.12 Complexity of order picking system (Source: de Koster et al., 2007) . . . . . . . . 48 3.13 Outline of the theoretical framework, for selection of warehouse automation sys- tem(s) for component storage - in a multiple assembly line manufacturing context, that is developed in this section. . . . . . . . . . . . . . . . . . . . . . . . . . . 50 4.1 Approximate production layout of the case facility . . . . . . . . . . . . . . . . . 53 4.2 The flow rack system, which is the main storage system used for storing load car- riers within the case facility. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 55 4.3 Visualizes the flow principle in the flow rack system used at the case facility. . . . 55 4.4 Shows the route of the continuously going tug-train “V1”, which supplies the Centralized storage/Kitting area with components being stored at other locations within the facility. . . . . . . . . . . . . . . . . . . . . . . . . . . . 59 5.1 Application of AS/RS based on number of SKUs vs Picking rate (Source: Manzini, 2012; Novara, 2020) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 78 5.2 Different configuration of LHDs within Dematic miniload AS/RS (source: demtatic, n.d) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 79 5.3 Swisslog Tornado - miniload (source: Swisslog 1, n.d) . . . . . . . . . . . . . . . 80 5.4 Swisslog Cyclone- shuttle based (source: Swisslog 2, n.d) . . . . . . . . . . . . . . 81 5.5 Kardex Remstar horizontal carousel - L shaped station (source: Kardex 1, n.d) . . 82 xv List of Figures 5.6 Kardex Remstar Megamat RS- Vertical carousel (source: Kardex 2, n.d) . . . . . . 83 5.7 Shuttle XP - Vertical Lift Module (source: Kardex 4, n.d) . . . . . . . . . . . . . 84 6.1 The theoretical framework, developed in section 3.4, that is applied on the empirical data in the analysis section 6.1 and 6.2 . . . . . . . . . . . . . . . . . . . . . . 85 6.2 Facility floor areas, which are available for installing warehouse automation sys- tem(s). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 87 6.3 The blue dashed arrows show the transport corridors, where all material flows pass and which cannot be blocked by warehouse automation system(s) . . . . . . . . . . 90 6.4 The relationship between end customer demand characteristics, the assembly lines’ demand characteristics and the requirements on the internal material supply pro- cesses (including any warehouse automation system(s)). . . . . . . . . . . . . . . 102 6.5 The incoming components to the warehouse automation system(s) are transported in groups of load carriers, while outgoing components are transported in groups of kits, load carriers or individual components depending on the storage policy and material feeding principle applied. . . . . . . . . . . . . . . . . . . . . . . . . . 106 6.6 The relationship between the order picking/kitting cycle time, for the AAA:1, AAA:2, BBB:1, CCC:1,and DDD:1 assembly line and the takt time at the re- spective assembly line. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 114 6.7 For the AAA:3 and BBB:2 assembly line, which has no dedicated component pick- ing/kitting operator, the component picking/kitting is a work task that is part of the takt time at the respective assembly line. . . . . . . . . . . . . . . . . . . . . 115 6.8 A visualisation of the inter-dependencies between the recommendations about what warehouse automation system(s) to use, which material feeding principle to apply for each assembly line, what storage policies to apply , and where to locate the warehouse automation system(s). . . . . . . . . . . . . . . . . . . . . . . . . . . 125 6.9 Approximate production layout associated with the main recommendations. The recommended new common kitting area for the AAA:1, AAA:2, AAA:3, CCC:1 and DDD:1 will be located at the existing Centralized storage/Kitting area and will house five vertical carousels in total. The two vertical carousels that are dedicated for the BBB:1 respectively the BBB:2 will be located next to these two assembly lines (e.g., in the purple marked area labeled V.C.s BBB). . . . . . . . . . . . . . 127 7.1 The Theoretical Framework (see Chapter 3), which is applied on the Empirical data (see Chapter 4) in the Analysis (see Chapter 6). . . . . . . . . . . . . . . . . . . 145 xvi List of Tables 1.1 Characteristics of the case company´s four main product groups assembled within the case facility. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3 3.1 Main characteristics described by Jonsson and Mattsson (2009) of five different manufacturing strategies. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15 3.2 Summary of implications described by Slack and Lewis (2015) for an operation with High respective Low level of the Four Vs . . . . . . . . . . . . . . . . . . . . . . 16 3.3 Descriptions and examples of benefits with the five generic performance ob- jectives described by Slack and Lewis (2011). . . . . . . . . . . . . . . . . . 17 3.4 Provides examples of risk factors provided in ISO 11228, which are associated with manual material handling. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20 3.5 Description of the six criteria a performance measurement system should fulfill, according to Caplice and Sheffi (1995). . . . . . . . . . . . . . . . . . . . . . . . 22 3.6 Examples of three types of manufacturing logistics performance measurements de- scribed by Jonsson and Mattsson (2009). . . . . . . . . . . . . . . . . . . . . . . 23 3.7 Description of a selected number of performance measures within manufacturing logistics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24 3.8 Descriptions and Implications described by Goldsby and Martichenko (2005) for seven sources of logistics waste. . . . . . . . . . . . . . . . . . . . . . . . . . . 30 4.1 Summary of the main characteristics of the four main product groups, which are assembled in the case facility. . . . . . . . . . . . . . . . . . . . . . . . . . . . 52 4.2 The total number of unit loads, which existed within the case facility at a specific date during the year 2020. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 57 4.3 The percentage distribution of primary storage locations and secondary storage locations, at a specific date during the year 2020. . . . . . . . . . . . . . . . . . 57 4.4 Summarizes the main characteristics of the AAA:1, AAA:2, and AAA:3 assembly line, in which AAA products are assembled. See Appendix Q for calculations of production pace. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 62 4.5 Summarizes the main characteristics of the BBB:1 and BBB:2 assembly line, in which BBB products are assembled. See Appendix Q for calculations of production pace. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 66 4.6 Summarizes the main characteristics of the CCC:1 assembly line, in which CCC are assembled. See Appendix Q for calculations of production pace. . . 69 4.7 Summarizes the main characteristics of the DDD:1 assembly line, in which DDD products are assembled. See Appendix Q for calculations of production pace. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 71 4.8 Summarizes the logistics performance measures and the ergonomic influ- encing variables used by the case company. . . . . . . . . . . . . . . . . . . . 74 xvii List of Tables 5.1 Warehouse automation systems (i.e., AS/RS) and supplier studied. . . . . . 77 6.1 How the number of handling tasks, per load carrier/stored component, differs be- tween possible combinations of storage policies and material feeding principles applied. 89 6.2 Summary of benefits and drawbacks with the storage policy Alternative 1. . . . . . 92 6.3 Summary of benefits and drawbacks with the storage policy Alternative 2. . . . . . 93 6.4 Summary of benefits and drawbacks with the storage policy Alternative 3 . . . . . 95 6.5 Summary of benefits and drawbacks with the storage policy-alternative 4. . . . . . 98 6.6 Summary of benefits and drawbacks with the storage policy Alternative PFSL . . . 100 6.7 Requirements for the Volume of finished products assembled, Volume of line items picked, and Volume of load carriers handled, for each of the seven assembly lines. The underlying calculations are available in Appendix O - Appendix Q. Some of the data presented in this table, are visualized in Appendix A - Appendix N. . . . . 104 6.8 Variation in demand per production day, for each assembly line. The underlying calculations are available in Appendix O - Appendix Q. Some of the data presented in this table, are visualized in Appendix A - Appendix N. . . . . . . . . . . . . . . 108 6.9 Maximum storage footprint for load carrier . . . . . . . . . . . . . . . . . . . . 109 6.10 Distribution of overall weight and number of bins between different types of load carrier, during a specific date in year 2020 . . . . . . . . . . . . . . . . . . . . . 110 6.11 Distribution of number of load carrier for purchase components . . . . . . . . . . 110 6.12 Distribution of number of load carrier for in-house manufactured components . . . 111 6.13 Summary of variety in demand, in the form of the handled load carriers’ technical characteristics, for each assembly line. . . . . . . . . . . . . . . . . . . . . . . . 111 6.14 Data of batch sizes, for each assembly line, which have implications for warehouse automation system(s) capacity requirements regarding the capability to present mul- tiple load carriers containing the same line item. The underlying calculations are available in Appendix O-Q . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 113 6.15 Data of takt time, production pace, minimum cycle time, and maximum production pace that have been calculated for each assembly line by using annual production data, for year 2020, extracted from the case company’s ERP system. (See Appendix O-Q for the underlying calculations.) . . . . . . . . . . . . . . . . . . . . . . . . 115 6.16 A summary of how high performance in the selected warehouse automation sys- tem(s) influence the performance of the main logistics KPIs used at the case com- pany. That is, high process performance considering a performance measurement being presented in the far right column, have positive effects on the performance of one, or several, of the main logistics KPIs being presented in the middle column. . 124 6.17 The primary characteristics of the main recommendations - for a new internal material supply operation. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 126 6.18 Summarizes the strengths and weaknesses related to how well the requirements, which are defined in the first part of the Analysis (see Section 6.1), are met by the main recommendations. Accordingly, this table also summarizes the consequences for the internal material supply operation related to selecting the recommended Warehouse Automation Systems for component storage . . . . . . . . . . . . . . 133 xviii 1 Introduction In this chapter, this master’s thesis background, aim, limitations and research questions are presented. First, the thesis background is described in a wider industry context. Thereafter, is the background of the case company’s specific situation presented followed by a clarification about how the case company’s specific situation is related to the previously described wider context. After that, the aim, limitations and defined research questions of this thesis is presented. Finally, the issue of ethical, societal and ecological aspects are brought up. 1.1 Background The trend of increased globalization of industrial markets not only provides opportuni- ties for companies in the form of increased sales potential, but it also stiffens the market competition (Rüttimann, 2018). A company’s competitive advantage derives from all its performed activities, which make activities the basic unit of competitive advantage (Porter, 1998). The part of a company that creates or delivers its products and services is referred to as Operations (Slack & Lewis, 2015). Since all organizations in one way or another try to add value by producing products and/or services, either for internal or external customers, operations is a function that exists within all companies even though it internally may be called something else (Slack & Lewis, 2015). Logistics is one such operation that supports the core value-adding processes. Logistics operations can be divided into internal and external logistics, where internal logistics deals with the flow of material within a facility (Groover, 2008; Gupta & Dutta, 1994). Thus, internal logistics includes areas of material supply and material management such as goods receiving, material handling, material feeding, storage and inventory, internal transport and packaging (Granlund, 2014). Even though internal logistics activities in general are considered as a non-value adding activity, it plays an important role in the overall function of organization in terms of business goals (Granlund, 2014). For example, it affects a com- pany’s readiness to deliver in a business environment that is ever more characterized by an increasing variety of products, higher ordering frequency of smaller lots, and tougher requirements for shorter delivery times (ten Hompel & Schmidt, 2007). In addition, a ma- jor reason for internal stock keeping is to ensure the productivity of expensive production processes by securing the material supply, for example, by buffering semi finished products between processes that differ regarding process time, output quantities, etc. (ten Hompel & Schmidt, 2007). Further, the increased market demand requires an increased focus on internal logistics since it supports operational performance of, for example, efficient mate- rial handling solutions and just-in-time supply of materials (Granlund, 2014). Hence the trend of increasing market demand creates a huge pressure and importance of logistics operations (Kartnig et. al, 2012). 1 1. Introduction Operational effectiveness refers to a company´s relative performance of activities, which are the same or similar as the activities performed by the competitors (Porter, 1996). Even though operational effectiveness is not sufficient to reach or sustain superior performance in relation to competitors, it is a prerequisite for it (Porter, 1996). A company can improve its operational effectiveness, for example by eliminating wasted efforts and implementing more advanced technology, thereby influencing the level of differentiation and its cost po- sition in relation to competitors (Porter, 1996). Though, purchasing top notch technology is not a guarantee for a company to reach or sustain superior performance, instead this is enabled by the ability to perform activities better than what competitors do (Hayes & Upton, 1998). Automation is one such way to improve competitiveness in operation. For example within the manufacturing industry, automation is known as a means of improving efficiency, pro- ductivity, quality and safety as well as lowering costs (Cruz Di Palma and Basaldúa, 2009; Frohm et al., 2006; Groover, 2008; Michalos et al., 2010). Automation has registered a huge mark on the operational areas such as picking, packaging, labeling, replenishment, storage and retrieval (McCrea, 2020). A survey conducted by Peerless Research Group (PRG) in 2020, indicates that there is an increase in respondents who want to increase the level of automation within those areas as compared to 2019 (McCrea, 2020). The performance of an operation can be measured by comparing how well it reaches the five generic performance objectives (i.e., Quality, Speed, Dependability, Flexibility, and Cost), which both directly and indirectly relates to the fulfillment of customer re- quirements (Slack & Lewis, 2011). Mediocre operations are usually easy to replicate by competitors, though matching an outstandingly performed operation is often very difficult (Hayes & Upton, 1998). A contributing factor for this, is that management of an opera- tion according to Slack and Lewis (2015) involves handling two interacting sets of issues, namely Processes and Resources. Operations and processes not only vary in terms of what technologies and skills that are needed to perform them, but also regarding the characteristics of the demand of the pro- duced products and services (Slack & Lewis, 2015). The Four Vs (i.e., Volume, Variety, Variation, Visibility) are considered as four influential characteristics of demand, which each one has implications for how to successfully manage a process (Slack & Lewis, 2015). When several different products or services, with both different technical and demand characteristics, must be processed within the same operation different implications arise for how to design and manage the operation in the best way. Some previous research indicates that the major problems with automation are not with the actual automation level or a lack of technology, but rather regarding the implementation and difficulty in selection of appropriate automation and the integration of it (Durrani et al., 1998; Sam- basivarao & Deshmukh, 1995). The success of automation depends on selecting, acquiring and implementing the right type of automation in consideration to the company’s needs, goals and prerequisites (Baines, 2004; Ceroni, 2009; Daim and Kocaoglu, 2008; Spath, Braun, & Bauer, 2009; Säfsten, Winroth, & Stahre, 2007). At the case company, a manufacturer of hydraulic pumps and motors, the management wants to improve the overall operational effectiveness by increasing the level of automa- 2 1. Introduction tion of the internal material supply operation. These plans are complicated by the fact that there are seven separate assembly lines, within the same production facility, which produces finished products that are assembled by components that differ regarding both technical features (e.g., weight and size) and demand characteristics (i.e, Volume, Variety, Variation, and Visibility). 1.1.1 The case company’s situation The case company is a manufacturer of hydraulic pumps and motors for mobile applica- tions. The company’s main customer base consists of manufacturers of heavy machinery and vehicles, from all around the world. This includes manufacturers of trucks, construc- tion machinery and forest machines. The case company’s product portfolio consists of four main product groups: AAA, BBB, CCC and DDD. Each main product group has a certain function and application area and includes multiple models, which in turn are made in several variants. The finished prod- uct’s weight is between 4.7 kg and about 100 kg. The large variety of finished products entails that there also is a large difference in the required number of input components between different product models and variants. At the case facility, the case company both assembles and tests all manufactured products before delivery. Also, a large share of the input components are manufactured in-house, though input components are also delivered to the case company as finished for assembling. In addition, some in-house manufactured components are shipped to external suppliers for further treatment, before being sent back to the facility. The goods received have the same manual process for all four main product groups. The case company requests suppliers to ship components in standardized 300x400mm bins, which are owned and provided by the case company. Though, some major suppliers do not fulfill this request and instead ship components in packaging of other sizes. Since the inventory rack system at the case company is adapted to the standardized bins, the case company needs to manually repackage some components to the 300x400mm bins. At the case facility, there are seven separate assembly lines in total for the four main product groups, which differ regarding volume assembled, assembly complexity, practiced internal material supply methods, etc. The characteristics of the four main product groups are described in Table 1.1. Volume assembled refers to the annual number of finished prod- ucts assembled, while assembly complexity refers to the number of line items per variant and the number of sellable variants. Table 1.1: Characteristics of the case company´s four main product groups assembled within the case facility. Main product group / Characteristics AAA BBB CCC DDD Volume assembled (annual number of finished products assembled) High ∼[40 000, 100 000] Medium ∼[20 000, 60 000] Low ∼[0, 20 000] Low ∼[0, 20 000] Assembly complexity A(number of line items) B(number of sellable variants) Medium A(∼33-40) B(∼1200) Low A(∼24-39) B(∼300) High A(∼62) B(∼50) Very high A(∼86) B(∼500) Number of assembly lines/assembly stations 3 2 1 1 3 1. Introduction The case company strives to store all components that are ready for assembling, both in-house manufactured and purchased finished components, in a centralized storage area that is located next to the assembly area. The implementation of this storage strategy has been constrained due to the limited storage floor space. In addition, the use of forklifts within the facility is forbidden due to safety concerns. As a result, all material handling is currently performed manually by personnel. The utilization rate of the total indoor height measuring about 7 m is remarkably low since ergonomic requirements limit the material racks´ height to only 1.6 meters above floor. In addition, some assembly lines are using material kitting as a material supply method which further restricts the height of racks, since the kitting is done manually by picker-to-parts method. Due to the fairly high complexity, the number of components in the product’s variants and limited floor space in the centralized storage area currently drives the case company to store components in multiple storage locations within the facility. This includes com- ponents that are both purchased as finished for assembling and in-house manufactured ones. Some in-house manufactured components are, due to constrained space availability, stored both near their respective pre-manufacturing cell and in additional storage loca- tions within the facility. The technical features (i.e. size and weight) of the components is one of the factors contributing to this. As a result of multiple storage locations, there is an increased internal material handling including transportation between those storage locations and transportation from those storage locations to the different assembly lines. Further, the seven assembly lines at the case company use different material feeding methods, in some cases a combination of sev- eral. This also entails increased internal transportation of components from one storage location to another, for instance, from storage locations in the pre-manufacturing area to the centralized storage area where some kitting takes place. These combined problems in the material supply operation, like multiple transportation, storage and handling of com- ponents along with the underutilized warehouse space (i.e. height), requires an extensive study in order to improve the overall operational effectiveness. The case company, views increased automation as a competitive advantage and is now in the initial phase of introducing an automated internal material supply operation. Cur- rently, the company is piloting a project with an Autonomous Mobile Robot that poten- tially may be used to transport components from the centralized storage area to different assembly lines. The case company sees large potential in better utilizing the available factory space (i.e., height), through increasing the level of automation in the centralized storage area and therefore wants to explore and evaluate implementing warehouse automa- tion system(s). Though, the case company recognizes the importance of a solution that does not sub-optimize the performance of the centralized storage area´s activities at the expense of the effectiveness of the actual material feeding processes. Implementing ware- house automation system(s), at the centralized storage area, will undoubtedly interfere with the currently practiced material supply methods (e.g. the milk-runs and component kitting procedures). Since a main objective with implementing warehouse automation system(s) at the case company is to make the extensive internal material handling more effective, an evaluation of the currently practiced material feeding principles´ compati- bility with any warehouse automation system(s) must be conducted. If the warehouse automation system(s) recommended are incompatible with currently practiced material feeding principles, suggestions for new feeding principles are required. 4 1. Introduction 1.1.2 The case company’s situation in relation to the wider context In order to improve the company’s competitiveness, the management at the case company plans to increase the level of automation of the internal material supply operation. These plans are complicated due to the fact that there are seven separate assembly lines within the production facility, which all assembles finished products that differ regarding both technical features (i.e., weight and size) and what Slack and Lewis (2015) refer to as de- mand characteristics (i.e., Volume, Variety, Variation, and Visibility). These differences are transformed into similar differences regarding the input components, which directly in- fluences the internal material supply operation of components and semi finished products. Further complexity is caused in the internal material handling by the case company´s practice of different material supply methods (e.g., kitting, pick-and-place, and milk-run) for different assembly lines. This complex situation, including a large variety in both technical features and demand characteristics of the input components complicates the case company´s selection of warehouse automation system(s) that reaches all performance objectives (i.e., Quality, Speed, Dependability, Flexibility, and Cost) for all seven assembly lines. A further complicating factor for the case company, is the lack of information about what to consider when increasing the level of automation of an internal material storage system, in the company’s multiple assembly line manufacturing context. 1.2 Aim The case company believes that increasing the level of automation of the internal material storage systems will support its competitiveness, but it is unclear for the company what type of warehouse automation system(s) that are suitable. A reason for this, is the lack of recommendations for what a manufacturing company in the case company’s multiple assembly line context should consider when selecting warehouse automation system(s) for component storage. The purpose of this master’s thesis is to explore how the case company can improve its internal materials supply operation, by increasing the level of automation of the compo- nent storage systems. This includes identifying and analysing important requirements that should be considered by the case company when selecting warehouse automation system(s) for component storage. Further, the purpose of this master’s thesis also includes to pro- vide recommendations for what type of warehouse automation system(s) for component storage the case company should select. In addition, the purpose includes identifying and analysing the consequences for the case company’s internal material supply operation of selecting the recommended warehouse automation system(s) for component storage. For example, this includes evaluating the compatibility between the case company’s currently practiced material feeding principles and the recommended warehouse automation sys- tem(s). Fulfilling the purpose of this master’s thesis will support the case company in developing a final detailed solution for an improved internal material supply operation, which involves a higher level automation of the component storage systems. In addition, this master’s thesis as a whole is expected to be useful also for other companies that operate in a simi- 5 1. Introduction lar context, which plan to increase the level of automation of their own internal material storage system. 1.3 Research questions The aim of this master´s thesis is to provide answers for the following research questions (RQs): RQ1: What are important requirements for the case company to consider when selecting warehouse automation system(s) for component storage? Answering RQ1 will contribute to the identification and understanding of performance requirements that should be considered when selecting warehouse automation system(s) for component storage, in order to secure a high performing solution when increasing the level of automation of the internal material supply operation. RQ2: What type of warehouse automation system(s) for component storage, are recom- mended for the case company to select? Answering RQ2 will contribute to the identification of what warehouse automation sys- tem(s) that are suitable for the case company to select for component storage, taking important performance requirements that were identified when answering RQ1 into con- sideration. RQ3: What are the consequences for the case company’s internal material supply opera- tion of selecting the recommended warehouse automation system(s) for component storage? Answering RQ3 will contribute to an understanding of how selection of the recommended warehouse automation system(s) for component storage will affect the case company’s cur- rent internal material supply operation, for example regarding the systems compatibility with different material feeding principles. 1.4 Limitations This master´s thesis does not aim to develop a single highly detailed and functional final solution of warehouse automation system(s) for component storage at the case company. Further, financial aspects like investment, operations, maintenance, and liquidation costs related to the selection of warehouse automation system(s) for component storage, are out of the scope of this thesis. In addition, material handling of finished products and raw materials (i.e., materials that will be pre-manufactured in-house before entering the assembly processes), are also out of the scope of this thesis. 6 1. Introduction 1.5 Sustainability - Ethical, societal and ecological aspects When conducting research, it is important to consider potential consequences for a sus- tainable development. Accordingly, potential effects on the sustainable development being caused by either the research methods applied or the expected research outcomes should be considered before initiating a research project. No negative effects related to sustainability (i.e., ethical, societal and ecological aspects), can been expected as a result of the implementation and/or the completion of this mas- ter’s thesis project. This means, neither the research methods applied (see Chapter 3 - Methodology) nor the answers provided for the asked research questions (see Chapter 7 - Conclusion), are expected to have any negative ethical, societal, or ecological conse- quences for any stakeholder or third party. Though, since the work environment for the operators within the case company’s facility will be effected by an increased automation of the component storage, the aspects of ergonomics has to be considered in the analysis. 7 1. Introduction 8 2 Methodology This chapter explains the methodology for how the aim of this master´s thesis has been fulfilled. A case study has been conducted to answer the research questions defined in Chapter 1 - Introduction. A case study is an inquiry-based research strategy that aims to describe, understand, predict, and/or control the individual research object, e.g. a process, organization, or industry (Woodside, 2010). A specific case study can include all, a combination, or just a single one, of these four types of research objectives and case study research can be used for both theory testing and theory building and is not limited to one specific set of research methods, instead, a combination of both qualitative and quantitative research methods within the same study may be advantageous (Woodside, 2010). Though, there exists no research method that is suitable to use in all case study contexts (Woodside, 2010). This case study is based on both qualitative and quantitative research methods. 2.1 Project procedure In order to get a deeper understanding of this thesis’ topic and to enable answering the research questions (RQs) defined, a comprehensive literature review was conducted in the early phase of the thesis project. This involved the study of existing literature and re- search papers, with focus on different aspects of internal material supply processes and warehouse automation systems. The findings from this literature review conducted, has been summarized in the Theoretical Framework (see Chapter 3). This Theoretical Frame- work forms the basis for the Empirical data collection (see Chapter 4) and the Analysis (see Chapter 6). A prerequisite for answering RQ1 (i.e., What are important requirements for the case com- pany to consider when selecting warehouse automation system(s) for component storage?), is to understand how the case company’s existing internal material supply operation is or- ganized, for example since this enables identifying inefficiencies that should be eliminated, or at least reduced, by future warehouse automation system(s) used for component storage. Further, to understand how the existing internal material supply operation is organized enables comparison between the current and a future state of the internal material supply operation. How the case company’s existing internal material supply operation is organized, has been mapped mainly by conducting interviews with different company stakeholders and through observations in the form of site visits. Examples of case company specific internal material supply operation characteristics that have been mapped, considers the currently applied: storage policies; material feeding principles; storage assignment policies; material replen- 9 2. Methodology ishment methods; packaging and unit loads; as well as what different types of handling and transport equipment that are used. In order to answer RQ1, empirical data corresponding to the different parts of the The- oretical Framework has been collected. Analysing this empirical data with support of the Theoretical Framework has allowed identifying and understanding important require- ments, for the selection of warehouse automation system(s) for component storage, which are imposed by the case company’s: Operations / Factory Environment (i.e., Produc- tion layout, Manufacturing strategies, Characteristics of demand, and Characteristics of assembly lines); Material Supply System (i.e., Materiel feeding principles, Handling Equip- ment, Storage systems, Packaging and unit loads, and Replenishment methods); as well as, Performance Objectives (i.e., Quality, Speed, Dependability, Flexibility, and Ergonomics). Answering RQ1 has allowed identification and understanding of performance requirements that should be considered by the case company when selecting warehouse automation sys- tem(s) for component storage, in order to secure a high performing solution. To provide recommendations for what type of warehouse automation system(s) for compo- nent storage that are recommended for the case company to select, requires identification and understanding of important requirements for such selection. This makes answering RQ1 (i.e., What are important requirements for the case company to consider when select- ing warehouse automation system(s) for component storage?) a prerequisite for answering RQ2 (i.e., What type of warehouse automation system(s) are recommended for the case company to select for component storage?) In order to answer RQ2, a market research has been conducted where the performance characteristics of different types of warehouse automation system(s), which are available on the market, have been mapped. These supplier model specific performance character- istics have been combined with general model performance characteristics data, found in the literature review, of different types of warehouse automation system(s). These findings about the performance characteristics of different types of warehouse automation systems, have then been compared to the requirements that are imposed by the case company’s Operations / Factory Environment, Material Supply System, and Performance Objectives, which are found in the answer provided for RQ1. Answering RQ2 has allowed identifi- cation of what warehouse automation system(s) for component storage that are suitable for the case company to select, taking important requirements that were identified when answering RQ1 into consideration. In order to answer RQ3 (i.e., What are the consequences for the case company’s internal material supply operation of selecting the recommended warehouse automation system(s) for component storage?), the strengths and weaknesses related to implementing the ware- house automation systems recommended, in the answer for RQ2, with respect to how well the performance objectives, defined in the answer for RQ1, are reached and how well the main recommendations are compatible with the requirements imposed by the Operation / Factory Environment and the Material Supply System, which also are defined in the answer for RQ1. Answering RQ3 has contributed to an understanding of how selection of the recommended warehouse automation system(s) for component storage, will affect the case company’s current internal material supply operation, for example regarding the systems compatibility with different material feeding principles. 10 2. Methodology 2.2 Empirical data collection In the Analysis (see Chapter 6), both qualitative and quantitative data are used. The non-numerical qualitative data has been collected through primary sources in the form of multiple structured, semi-structured, and unstructured stakeholder interviews, conducted with different managers within the case company, as well as by observations, through several site visits, where non-numerical data mainly about the existing material supply operation has been collected. To interview managers with different responsibility ar- eas within the case company, has brought different perspectives into the analysis, which strengthens the viability of the answers for the research questions. In addition, some managers have also been interviewed multiple times. Further, that several managers have been interviewed and partly been asked the same questions strengthens the reliability of the collected empirical data since answers perceived as divergent have been followed up and either rejected or confirmed by other interviewees. This has also reduced the risk of biased answers, from individual stakeholders, influencing the empirical data collection and in turn the Analysis. The numerical quantitative data has almost exclusively been collected through secondary sources in the form of the case company’s Enterprise Resource System (ERP system), Plan For Every Part files (PFEP files), and Bill-of-Materials (BOMs). The data extracted from these secondary sources has in the Analysis been combined in order to enable the numerical calculations required, for example regarding throughput requirements and storage volume capacity requirements imposed on the selection of warehouse automation system(s). The specific sources for each type of quantitative data is provided, where the focal data is presented in this thesis. Further, that almost exclusively secondary data has been used for the numerical calculations, increases the reliability and validity of the analysis. This since risks related to mistakes or errors in data collection and data transferring has been minimized. The qualitative and quantitative empirical data are used in an integrated approach in the analysis, where the theoretical framework is applied on all the empirical data collected. This further strengthens the validity of the answers for the defined research questions. 11 2. Methodology 12 3 Theoretical Framework Due to the lack of cohesive frameworks for the selection of warehouse automation sys- tem(s) for component storage, within a material feeding context where multiple assembly lines are the direct customers of the stored components, this chapter summarizes aspects of an internal material supply operation that, through reviewing literature and previous research, have been identified as important to consider when selecting warehouse automa- tion system(s) for component storage in the case company’s situation. This theoretical framework forms the basis for the empirical data collection (see Chapter 4) and for the Analysis (see Chapter 6). 3.1 Operations / Factory Environment In this section, first the main characteristics of the customer order decoupling point and related manufacturing strategies are described. Next, what implications the characteristics of demand have on a material supply process are provided. Then, different process per- formance objectives and the importance of fit between such objectives and the design of a production system are described, as well as, the importance of monitoring processes with the support of performance measurements. Here, the importance of making ergonomic considerations within manual material handling are brought up. Also, examples of im- portant performance measures within internal manufacturing logistics are provided. The following subsection describes how performance objectives can be reached by improving internal manufacturing logistics processes, through reducing process variability and apply- ing Lean Six Sigma logistics. Thereafter, several main characteristics of different assembly lines and what implications such differences have for material feeding processes are de- scribed. This is important for the scope of this thesis since there are seven assembly lines, with different main characteristics, within the case company’s facility being in focus. After that, different aspects of material supply strategies are described, followed by a description of the area of automation of internal logistics. 3.1.1 The Customer order decoupling point & Manufacturing strategies The case company applies different manufacturing strategies, which affect the internal ma- terial supply operation. It is important to consider the customer order decoupling point (CODP) when designing a manufacturing operation, both on the strategic, tactical and operational level (Olhager, 2010). What manufacturing strategy that is practiced within a company has implications for how to manage the material supply and production pro- cesses, for example when the degree of information about the final product is low at the customer order decoupling point the following processes must be more flexible than when the degree of information is high (Jonsson & Mattsson, 2009). 13 3. Theoretical Framework A manufacturing company can according to Jonsson and Mattsson (2009) be classified regarding how customer order initiated its operations are, in other words, how integrated its production functions are with the customer orders received. This means that the cus- tomer order point (i.e., the BOM-level from which the product design is customer order specific) or the similar customer order decoupling point (i.e., the BOM-level from which material supply and value-adding activities needs a customer order to be initiated) can be used for categorizing a manufacturing company (Jonsson & Mattsson, 2009). Before the customer order decoupling point all material supply and value-adding activities are initi- ated by forecasts, while afterward, customer orders received decide the initiation (Jonsson & Mattsson, 2009). From a supply perspective, the degree of order information according to Wikner and Rudberg (2005) concerns four mutually independent supply issues: What product does the customer want?; How much of that product is wanted?; When is delivery wanted?; and Where is delivery wanted?, which all must be specified to make a customer order fully defined. Since the production lead time often is longer than the delivery lead time demanded by customers (see Figure 3.1), decisions regarding the four aforementioned supply issues must be taken before full information is available about the demand require- ments (Wikner & Rudberg, 2005). In addition, there is no sure relationship between the level of information certainty between the four different supply issues (Wikner & Rudberg, 2005). Based on the customer order decoupling point five manufacturing strategies can be sep- arated (see Table 3.1), which all differ regarding: the level of customer order integration within material supply and manufacturing activities; where the stock point(s) is located within the material flow; and the degree of information at order receipt (Jonsson & Matts- son, 2009). Figure 3.2 shows the relationship between different manufacturing strategies and the CODP. Most companies produce a mix of make-to-stock and make-to-order prod- ucts, which affect the design of the production system (Olhager, 2010). In an assemble-to- order environment, it is necessary to design different parts of the supply chain according to either make-to-stock or make-to-order strategy (Olhager, 2010). Figure 3.1: The Customer Order Decoupling Point (CODP) in relation to production and delivery lead time. Adapted from Wikner and Rudberg (2005). 14 3. Theoretical Framework Figure 3.2: The Customer Order Decoupling Point (CODP) in relation to four manufacturing strategies. Adapted from Wikner and Rudberg (2005) Table 3.1: Main characteristics described by Jonsson and Mattsson (2009) of five different manufacturing strategies. Manufacturing Strategy The level of customer order integration within material supply and manufacturing activities The degree of information at order receipt Location of stock point(s) Engineer-to- order (ETO) All activities from product design, material supply, to final assembly is customer order initiated. Very low There is only one stock point, which is located between the procurement and fabrication processes. Make-to- order (MTO) Parts of the material supply and manufacturing activities involving input- components and products are executed based on forecasts, i.e., they are not customer order initiated. Low to average There is a stock point located both between the procurement and the fabrication process and within the fabrication process. Assemble-to- order (ATO) All material supply and pre-manufacturing activities of standardized input-components are executed based on forecasts, i.e., they are not customer order initiated. Only the final composition of input-components and the assembling activities are specified and initiated by customer orders received. Average to high There is a stock point located both between the procurement and the fabrication process and between the fabrication and assembly process. Make-to delivery-schedule (MTDS) Products are either completely standardized or customer specific for certain customers, though material supply and manufacturing activities are not customer order initiated but instead executed based on forecasts, stock levels, or a delivery schedule. High As for the ATO-strategy, there is a stock point located both between the procurement and the fabrication process and between the fabrication and assembly process. Make-to-stock (MTS) Product specifications are completely known before customer order is received and the company manufactures standardized finished products, which are stored in stock until a customer order initiates delivery. All material supply and manufacturing activities are as for MTDS executed based on forecasts, stock levels, or a delivery schedule. High In addition to the stock point locations within the MTDS-strategy, there is also a stock point between the assembly and delivery process. 15 3. Theoretical Framework 3.1.2 Characteristics of demand - influences material supply processes The case company’s product portfolio includes a large variety of products, which have implications for the assembly operation and consequently for the related material supply processes. Operations and processes not only vary in terms of what technologies and skills that are needed to perform them, but also regarding the characteristics of the demand of the produced products and services (Slack & Lewis, 2015). The Four Vs (i.e., Volume, Variety, Variation, and Visibility) are four influential characteristics of demand (see Table 3.2), which each one has implications for how to successfully manage a process (Slack & Lewis, 2015). Table 3.2: Summary of implications described by Slack and Lewis (2015) for an operation with High respective Low level of the Four Vs . Characteristics of demand (the Four Vs) Implications for an operation with High level of focal “V” (OBS: Low level for Volume) Implications for an operation with Low level of focal “V” (OBS: High level for Volume) Volume (i.e., demanded volume) Low degree of repetition, Labor intensive, Low level of specialization, Low grade of systematization, High cost per unit. High degree of repetition, Capital intensive, High level of specialization, High grade of systematization, Low cost per unit. Variety (i.e., variety in characteristics of products and services) High degree of flexibility, High level of complexity, Match customer needs, High cost per unit. Low degree of flexibility, Routine, High level of standardization, Regular, Low cost per unit Variation (i.e., variation in demanded volume) Varying capacity needs, Anticipation, High degree of flexibility, In touch with demand, High cost per unit. Stable capacity needs, Predictable, Routine, High utilization rate, Low cost per unit. Visibility (i.e., how exposed the operation´s value-adding is to customers) Short waiting tolerance, Customer perception is important, Customer contact skills are required, High cost per unit. Time discrepancy between production and consumption, High level of standardization, High utilization rate, Low cost per unit. HIGH UNIT COST LOW UNIT COST 3.1.3 Performance Objectives - fit production system design with man- ufacturing objectives It is crucial that implementation of warehouse automation system(s) improves the case company’s operational performance. According to Devaraj, Hollingworth, and Schroeder (2004) manufacturing performance is enhanced when manufacturing design choices fit with the manufacturing objectives. 3.1.3.1 Five generic performance objectives In practice, the five generic performance objectives (i.e. Quality, Speed, Dependability, Flexibility, and Cost), which are applicable to all operations, form the basis for a reason- ably well-defined set of required performance objectives (Slack & Lewis, 2011). All these five generic performance objectives are related to the underlying business task of fulfilling the customer needs (Slack & Lewis, 2011). These five performance objectives are often interlinked and even if any of them happens to be of low direct importance for meeting customer requirements, it may still be valuable for the operation´s overall performance 16 3. Theoretical Framework due to the internal benefits it creates (Slack & Lewis, 2011). In Table 3.3, descriptions and examples of internal and external benefits of the five generic performance objectives described by Slack and Lewis (2011) are presented. Table 3.3: Descriptions and examples of benefits with the five generic performance ob- jectives described by Slack and Lewis (2011). Generic performance objective Description Examples of potential internal benefits Examples of potential external benefits Quality Quality often refers to the specification of the produced products or services but can also refer to how well the produced products or services are fit for purpose. Quality is multidimensional and several aspects of specifications are required to describe the desired features of the produced products or services. In addition, conformance quality regards how reliably and consistently the focal operation itself reaches the defined product or service specifications. Fewer errors in processes, less complexity and disruption, higher internal reliability, less processing costs Fewer errors in products/services, highly specified products/services, increased reliability of products and services Speed Speed refers to elapsed time and in its most basic form concerns the time from start to the end of a process in an operation. It may involve external happenings, like the customer order lead time that may include both order-handling, core processing, queuing, delivery, installment, though speed may also refer, more narrowly, to only an internal process´ throughput time. Faster throughput times, less inventory, less overhead costs, less processing costs Shorter delivery times, faster response to customer requests Continued on next page 17 3. Theoretical Framework Continuation of Table 3.3 Flexibility Flexibility has two different meanings: the process´ range flexibility, e.g. the ability to produce different variants of products or services, or produce with varying output levels; and the process´ response flexibility, i.e. how fast, smoothly, and cheaply the process can change between different production states. Examples of different kinds of flexibility, which may influence an operations competitiveness are: product and service flexibility, i.e. how easily modifications to existing products/services are made and how easily novel products are introduced; mix flexibility, i.e. how easily the produced variety of products/services is changed; volume flexibility, i.e. how easily the aggregated output level is changed; and delivery flexibility, i.e. how easily planned delivery dates are changed. Increased responsiveness to unforeseen events, increased responsiveness to required changes in activities, less processing costs Frequent new product/service introductions, wide range of products/services, adjustable volumes, adjustable deliveries Continued on next page 18 3. Theoretical Framework Continuation of Table 3.3 Dependability Dependability refers to keeping delivery promises regarding the delivery time, i.e. deviations between due delivery time and actual delivery time, which often makes it linked with the speed performance objective. In addition, fast throughput times often facilitate high dependability due to the related increased level of process control. Increased trust in the operation, fewer contingency plans required, increased internal stability, less processing costs Increased on-time delivery to customers Cost Cost refers to any financial input that is required for enabling the focal operation to produce the assigned products or services. In general, an operation requires: operating expenditures; capital expenditures; and working capital. higher margins Lower prices End of Table 3.3 3.1.3.2 Ergonomics - the importance of human well-being for high performing material handling Since any changes to the case company’s internal material supply operation will affect the employees work conditions, ergonomic aspects must be consider when selecting warehouse automation system(s) for component storage. Internal logistics activities involve both movement of materials as well as handling of equipment and the discipline of ergonomics is used to improve the execution of such activities, by simultaneously acknowledging the importance of both good human and physical work conditions as well as effective and efficient performance (Loos, Merino & Rodriguez, 2016). For a company to reach its or- ganizational objectives, it is crucial to create conditions that support human well-being, followingly, ergonomic factors must be taken into consideration when designing processes (Loos, Merino & Rodriguez, 2016). This includes evaluating both explicit physical loading and psycho-social aspects, with support of defined ergonomic goals and related indicators (Neumann, Winkel, Medbo, Magneberg & Mathiassen, 2006). Further, to consider er- gonomic factors early in the design phase of a production system have a major impact on the ergonomic quality of the final system (Neumann, Kihlberg, Medbo, Mathissen, & Winkel, 2002). For example, when re-designing a production system through increasing the level of automation, it is important to consider both what work tasks should be re- 19 3. Theoretical Framework moved as well as the remaining work tasks since also the removal of specific repetitive monotonous activities, which are unwanted from an ergonomic perspective, can reduce operators overall load variation and consequently increase the workload concentration to fewer body parts (Neumann et al., 2002). Main ergonomic risk factors within manual material handling, which according to Rossi, Bertolini, Fenaroli, Marciano and Alberti (2013) all are related to fatigue and physical injuries, are: static or awkward work pos- tures; repetitive motions; forceful exertions; and pressure points. The International Organization for Standardization (ISO) provides a three-part frame- work (i.e., ISO 11228) for ergonomic evaluation of manual material handling, with the primary objective of making design improvements to manual handling operations (Rossi et al., 2013; ISO, 2003; ISO, 2007a; ISO, 2007b): 1. The first part of ISO 11228 provides ergonomic recommendations for manual lift- ing and carrying of objects with a weight of at least 3 kg and for normal walking speeds between 0.5-1.0 m/s on plane surfaces (ISO, 2003). Related risk factors are presented in Table 3.4. An ideal posture for manual handling is to: stand upright and symmetrically; keep a horizontal distance of less than 25 cm between the centre of mass of the operator and the object handled; and keep a vertical distance of less than 25 cm between the hand grip and the knuckle height (ISO, 2003). 2. The second part of ISO 11228 provides risk-assessment methods for whole-body man- ual pushing and pulling, as well as proposes limitations for such work tasks (ISO, 2007a). This includes tight restrictions of manual pushing and pulling of heavy objects, which can be avoided by automation, mechanisation, or through design adaptations of the work task or workplace (ISO, 2007a). Risk factors related to manual pushing an pulling are presented in Table 3.4. 3. The third part of ISO 11228 provides a risk-assessment model for high frequent repetitive work tasks involving manual handling of low loads, as well as proposes restrictions for such work tasks (ISO, 2007b). Examples of factors that, alone or in combination, are associated with ergonomic risks related to handling of low loads at high frequency are presented in Table 3.4. Such risk factors can be reduced by automation, mechanisation, work task enlargements, or job rotation (ISO, 2007b). Table 3.4: Provides examples of risk factors provided in ISO 11228, which are associated with manual material handling. Work tasks Examples of risk factors (alone, or in combination) Lifting and Carrying object position, working posture, mass and size of objects handled, frequency of execution, and exposure duration, etc (ISO, 2003). Pushing and Pulling used force, working posture, frequency of execution, exposure duration, distance, object characteristics, environmental condition, individual characteristics, work organization, etc (ISO, 2007a). Handling of low loads at high frequency working posture and movement of body parts, used force, exposure duration, frequency of executions, training level, quality and precision requirements on output, work organization, etc (ISO, 2007b). An additional positive effect of improving ergonomic work conditions for material han- 20 3. Theoretical Framework dling operators, is that by removing work tasks that involve heavy lifting, which excludes many potential employees, a company can broaden the possible worker pool as well as improving productivity and quality by reducing worker fatigue (Baudin, 2002). Further, since the preferable work height from an ergonomic point of view is between the shoulders and hip, which is a range that differs between individuals, the height of workstations or used fixtures should preferably be adjustable (Baudin, 2002). In addition, the material ex- posure at an assembly line strongly affects ergonomic conditions, for example, using small containers for material exposure reduces the ergonomic risk factors of shoulder elevation and trunk flexion (Finnsgård, Wänström, Medbo & Neumann, 2011). 3.1.4 Performance measurement systems - monitoring performance within manufacturing logistics The case company has a flow oriented production system, which according to Stricker, Echsler Minguillon, and Lanza (2017) generally are prone to disruptions, and especially lean practices like inventory, time and capacity buffer reductions make such a system´s performance immediately affected by such disruptions. Further, the probability of disrup- tions increases due to the general growing range of product variants and the increasing distribution of production processes (Stricker et al., 2017). Consequently, the costs of disturbances are rising due to the increasing need for process control and flexibility, which in turn raises reliability and consistency requirements on utilized equipment (Jonsson, 2000). In order to monitor and detect changes in a production system´s performance a well-developed multi-dimensional performance measurement system is needed (Stricker et al., 2017). To measure and follow-up the performance of manufacturing logistics processes enables a company to take adequate actions that allows it to reach performance objectives determined (Jonsson Mattsson, 2009). In addition, performance measures can be used for sizing the logistics system and for delegating responsibilities for logistics activities (Jons- son & Mattsson, 2009). 3.1.4.1 Criteria for a performance measurement system A performance measurement system should be process oriented and designed with cus- tomer value-adding in focus and should be designed so it highlights improvement potential and increases the effectiveness and efficiency of actions (Dörnhöfer, Schröder, Günther, 2016). Process performance indicators are seldom independent, instead, they are com- monly in some kind of conflicting or complementary relationship with one another (Kueng, 2000). It is important that there exists a balance between performance measures for indi- vidual process steps and performance measures that covers the whole process (Dörnhöfer et al., 2016). Since logistics often involves a complex set of interlinked activities there is according to Caplice and Sheffi (1995) a need for several complementary and supportive performance metrics to provide the management a comprehensive view of the logistics performance. Such performance measurements should support the company’s competi- tiveness and be aligned with its objectives and the overall business strategy (Jonsson & Mattsson, 2009). It is also important that a well-balanced set of financial and non-financial measures are used (Gunasekaran, Patel, & McGaughey, 2004). A well designed performance measurement system that is both complementary, comprehensive, and cohesive, in which performance 21 3. Theoretical Framework metrics are evaluated both individually and on a system-wide level, will improve the management’s decision making (Caplice & Sheffi, 1995). A performance measurement system should fulfill all six criteria that are described in Table 3.5 (Caplice & Sheffi, 1995). Table 3.5: Description of the six criteria a performance measurement system should fulfill, according to Caplice and Sheffi (1995). Criteria Description Comprehensive all relevant stakeholders for the process should be captured. Causally Oriented activities and indicators that affect both the current and future performance should be tracked. Vertically Integrated the overall business strategy should be translated to all relevant decision makers in the company. Horizontally Integrated all relevant activities, functions and departments along the process should be included. Internally Comparable potential trade-offs between different performance aspects (not between evaluation criteria) should be recognized and allowed. Useful decision makers should be able to understand and use the performance measurement system as a tool for decision making. 3.1.4.2 Three types of logistics performance variables Logistics performance variables can regarding their influence be divided into three types (see Table 3.6), i.e.: Revenue influencing logistics variables; Cost influencing logistics vari- ables; and Asset influencing logistics variables (Jonsson & Mattsson, 2009). The largest part of a manufacturing company’s current assets often consist of tied-up capital in raw- material, semi-finished items, and purchased components located both in different inven- tories and within the actual material flow, though, also accounts receivable is affected by the logistics procedures through the company’s delivery capacity, for example by the share of customer orders that are delivered in full (Jonsson & Mattsson, 2009). Since tied-up capital costs significantly influence a company’s profitability, in terms of return on capital employed (ROCE), the logistics performance of a company has a major impact on the overall financial results (Jonsson & Mattsson, 2009). More detailed descriptions of a selected number of revenue, cost, and asset influencing logistics variables, used within manufacturing logistics, are provided in the next subsection. 22 3. Theoretical Framework Table 3.6: Examples of three types of manufacturing logistics performance measurements de- scribed by Jonsson and Mattsson (2009). Performance measure Description Examples Revenue influencing logistics variables Includes customer-oriented performance variables. E.g., delivery precision, delivery reliability, delivery lead time, stock service level and flexibility. Cost influencing logistics variables Includes performance variables related to both production and the material flow. -Production (e.g., capacity costs, costs for changing rate of production, and set-up costs). -Material flow (e.g., transportation and handling costs, packing costs, inventory carrying costs, shortage costs and delay costs, and administrative costs) Asset influencing logistics variables Includes performance variables related to both fixed and current assets. -Fixed assets (e.g., the rate of capacity utilization, i.e., the quota between the produced volume and the nominal capacity). -Current assets (e.g., share of full deliveries, the amount of tied-up capital in monetary values, inventory turnover rate, and run-out time). 3.1.4.3 Key performance indicators within manufacturing logistics The case company uses several key performance indicators (i.e. measures), to control the assembly operation and related material supply processes. Within logistics, it is crucial to use logistics performance indicators to quantify the current state and potential im- provements (Dörnhöfer et al, 2016). The choice of performance measures is crucial for a company’s success since these affect: planning and control, on both strategic, tactical and operational level; evaluation of performance; and what future actions that should be taken (Gunasekaran et al., 2004). All companies have individual performance measure- ment needs and the chosen set of performance measures must reflect a company’s unique operations and its business context (Gunasekaran et al., 2004). The number of perfor- mance measures a company uses tend to accumulate over time, though generally it is better to use a few performance measures that cover areas that are most important for the company’s success (Gunasekaran et al., 2004). Financial performance measures are often valuable for external reporting and when making strategic decisions, while non-financial performance measures generally are more valuable for the control of day-to-day manufacturing operations (Gunasekaran et al., 2004). The performance of production operations has a great influence on both quality, flexibility, speed of delivery, delivery reliability, and product cost (Gunasekaran et al., 2004). In addition, the use of non-financial performance measures has a statistically significant direct positive effect on company profitability, but it also has a statistically significant mediating function for the indirect positive effects that lean management practices have on company profitability (Fullerton &Wempe, 2009). Descriptions of a selected number of performance measures used within manufacturing logistics, are presented in Table 3.7. 23 3. Theoretical Framework Table 3.7: Description of a selected number of performance measures within manufacturing logistics Performance measure Definition Relevance & Implications Order lead time The time elapsed between a customer makes an order and the customer receives the ordered product or service (Gunasekaran et al., 2004). Is a source of competitive advantage since it affects a company’s response time for meeting customer needs (Gunasekaran et al., 2004). On-time parts availability / On-time-in-full (OTIF) The right part should be delivered at the right time, in the right quality, in the right location, and in the right packaging (Dörnhöfer et al., 2016). On-time parts availability is crucial for ensuring high capacity utilization of an assembly line (Dörnhöfer, et al., 2016). Productivity / Throughput Is defined as the quota between outputs and inputs used, which refers to effectiveness of a system, where inputs are resources such as time, labor, or money, while examples of outputs are line items, the number of finished orders, or transactions (Park, 2012). Throughput decreases: as queue capacity between work stations decreases; as product and process variation increases; and as the number of sequenced work stations increases (Crandell & Burwell, 1993). Productivity may imply wrong decisions and behaviors if used in isolation (Goldsby & Martichenko, 2005). Capacity utilization / Utilization rate Can be defined as the ratio between produced volume and the nominal capacity, where nominal capacity refers to the capacity that is available under normal conditions (Jonsson & Mattsson, 2009). Has direct impact on a company’s response time for meeting customer needs since it influences lead time, flexibility, and deliverability (Dörnhöfer et al., 2016). Decreases as: que capacity decreases; the number of workstations increases; and product and process variation increases (Crandell & Burwell, 1993). Influences both the average and the variability of lead time through factory, see Figure 3.3 (New, 1993). May imply wrong decisions and behaviors if used in isolation (Goldsby & Martichenko, 2005). Cycle time deviation Cycle time is the time available, at each station, for each work cycle (Boysen, Fliedner, & Scholl, 2008). In other words, the time span between a workpiece´s two entries into two successive workstations (Boysen et al., 2008). If the cumulated task time for any station (i.e. the station time), in a paced line, exceeds the fixed common cycle time that work station becomes a bottleneck and the fixed common cycle time is not feasible (Boysen et al., 2008). If the cumulated task time for any station is shorter than the fixed common cycle time that workstation has idle time (Boysen et al., 2008). Takt time deviation Takt time is a measurement of production pace, which relates available production time to customer demand (Linck & Cochran, 1999). It determines the time available, at each workstation, for producing a single product, and is calculated as the ratio between available production time and the average customer demand per time period (Linck & Cochran, 1999), Deviation from takt time leads to either overproduction or underproduction in relation to customer demand (Linck & Cochran, 1999). Logistics processes should be timely in line with the manufacturing lines takt time (Dörnhöfer et al., 2016) Line fill-rate A measure of the probability that an assembly line will not have any disruptions, between successive component-storage replenishments, that is caused by component shortage (Bukchin & Meller, 2005). To maximize the line fill-rate is a main goal when designing an assembly line operation since it equals minimizing the probability of component shortage, which causes line stoppages (Bukchin & Meller, 2005). Large positive financial effects can be caused by even small increases in line fill-rate, since a single component shortage may cause major disruptions and idleness costs (Bukchin & Meller, 2005). Line efficiency A measure of an assembly line´s productivity that is calculated by the ratio between the sum of the cumulated task time , i.e. station time, for all workstations (t,sum) and the product of the number of station (n) and the common cycle time (c), i.e. Line efficiency = t,sum/(nc) (Boysen et al., 2007). Is used for evaluating the quality of the assembly line balance, by measuring the productive fraction of the total operating time (Boysen et al., 2007). Throughput efficiency Measures productivity of time and is defined as the ratio between work content and the elapsed time taken (New, 1993). Does not measure efficiency of the actual work tasks, though is measured for a given set of work methods and shift patterns (New, 1993). On item level, throughput efficiency = 1-Balance Delay loss, where Balance Delay loss equals (cycle time x number of stations - work content of unit)/ (cycle time x number of stations) (New, 1993). High levels can lead to: reduced customer order lead time in make-to-order environments; shorter forecast horizons and consequently the reduced exposure to forecast errors in make-to-stock environments; lower WIP-inventory levels; improved responsiveness to changes in demand; improved service levels with lower inventory levels; reduce a company’s risk exposure; and facilitate timely introductions of new products; (New, 1993). Though, improved throughput efficiency of an individual process does not automatically lead to improved overall company throughput efficiency (New, 1993). Work in Process (WIP) Inventory of unfinished products that are located in and between production processes (Jonsson & Mattsson, 2009). WIP provides buffers against process variability, which facilitate high machine utilization rates (Crandell & Burwell, 1993). Reduced average WIP may reduce throughput, i.e, the number of units processed per time unit (Crandell & Burwell, 1993). Increased WIP increases both the average and the variability of lead time through factory, i.e. the sum of process and waiting time (New, 1993; Crandell & Burwell, 1993). 24 3. Theoretical Framework 3.1.4.4 Conflicting and complementary relationships between performance measures Figure 3.3 and Figure 3.4 show examples of conflicting and complementary relationships, which according to Kueng (2000) commonly exist between different performance measures. Figure 3.3, which summarizes findings by New (1993), shows that high throughput effi- ciency shortens customer lead times, in a make-to-order environment, though it reduces the capacity utilization. In addition, Figure 3.3 shows that WIP-congestion causes capac- ity utilization, after a certain level, to fall (New, 1993). Figure 3.4 show that both average work-in-process, work pieces´ average time in system, average utilization, and average throughput increases as the queue capacity increases (Crandell & Burwell, 1993). The data in Figure 3.4 show that it is possible to both reduce work-in-process and shorten lead times, while preserving resource utilization and throughput, by reducing process variabil- ity (Crandell & Burwell, 1993). The importance of process variability for production sys- tems performance is further described in the next subsection about process improvements. Figure 3.3: Describes the relationship between Throughput efficiency, Capacity utilization and Lead time. Adapted from New (1993). 25 3. Theoretical Framework Figure 3.4: The relationship between Queue capacity and average work-in-process, average time in system, average utilization, and average throughput, respectively. Adapted from Crandell & Burwell (1993). 3.1.5 Process Improvements - reaching performance objectives It is important that future warehouse automation system(s) at the case company does not sub-optimize the component storage processes, instead, it should enable and facilitate ex- cellent performance of all interlinked processes, not at least the assembly processes. This subsection describes different aspects and concepts of how internal manufacturing logistics processes can be improved. 3.1.5.1 Designing a production system - three ways to handle process vari- ability A production system must be designed so internal variability in production throughput does not affect the company’s ability to satisfy customer needs (Mierzejewska, Castaneda- 26 3. Theoretical Framework Vega, Cochran, 2002). Such internal variability can be handled by: decoupling of pro- cesses, through implementing buffers in-between; excess capacity, by capital investments in equipment and labour; or by reduction of variation, which often is the most cost-efficient alternative even though it requires both skills and endurance (Mierzejewska et al., 2002). Buffers between processes in a production flow can be either visible, in the form of WIP material, or invisible, such as flexible distribution of work tasks, reduced worker concentra- tion (i.e. fewer operators in relation to the number of products), and assigning operators additional work tasks that are not directly linked to the production flow (Engström, Jon- sson & Medbo, 1996; Engström & Karlsso