IN
DEGREE PROJECT MATERIALS DESIGN AND ENGINEERING, SECOND CYCLE, 30 CREDITS
,
STOCKHOLM SWEDEN 2019
A Value Proposition for
Cloud-Enabled Process
Planning
SOFIA TOLLANDER
KTH ROYAL INSTITUTE OF TECHNOLOGY
KTH R
OYAL
I
NSTITUTE OF
T
ECHNOLOGY
M
ASTERT
HESISA Value Proposition for
Cloud-Enabled Process Planning
Author: Sofia TOLLANDER Supervisors: Dr. Vahid KALHORI Magnus LUNDGREN Master of Science
KTH School of Industrial Engineering and Management Department of Production Engineering
S-100 44 STOCKHOLM
Abstract
To stay competitive in today’s fast-paced market, manufacturing companies must shorten their time-to-market and decrease their costs by efficiently utilizing their resources. Here, improved software and better software integration throughout the product realization process is consid-ered to be a key enabler.
The aim of this thesis work has been to investigate the current workflow in design and process planning to outline a cloud-based application to support these activities. Pains and bottlenecks in these workflows have been identified through interview and field studies at six Swedish manufacturing companies of different sizes, in different industries, and with different opera-tional models.
The major areas of improvement were identified whereof one them, the initial activity of un-derstanding customers’ needs, was decided to further focus on. From receiving a request for quotation from the customer to acceptance of an order, the following time-consuming activi-ties were recognized: understand and discuss design intent as well as suggest possible design changes to improve manufacturability, analyze and review 2D drawings and 3D models, and develop order quotations.
In this thesis work, a mock-up prototype has been put forward. The intent with this is to bridge a gap that has been identified through the mapping between manufacturers needs and functionality of available CAD/CAM software and the identified areas of improvement from the workflow investigations.
The proposed solution, as presented in the mock-up prototype, has been validated together with three of the studied companies. At its current state, further improvements and validations are needed. Nevertheless, if further developed, it has the potential to create value within the entire manufacturing value chain.
Sammanfattning
För att vara konkurrenskraftig på dagens snabbrörliga marknad måste tillverkande företag minska sina tid till marknad och reducera sina kostnader genom att utnyttja sina resurser mer effektivt. Här anses förbättrad programvara och bättre mjukvaruintegration genom hela pro-duktrealiseringsprocessen vara en viktig möjliggörare.
Målet med detta arbete har varit att undersöka det aktuella arbetsflödet i design och process-planering för att föreslå en molnbaserad applikation för att stödja dessa aktiviteter. Problem och flaskhalsar i dessa arbetsflöden har identifierats genom intervjuer och fältstudier hos sex svenska tillverkande bolag i olika storlekar, i olika branscher och med olika operativa modeller. De viktigaste förbättringsområdena identifierades varav en, den initiala aktiviteten för att förstå kundernas behov, beslutades att ytterligare fokusera på. Från att ha mottagit en offertförfrågan från kunden till acceptans av en beställning kunde följande tidskrävande aktiviteter identi-fieras: förstå och diskutera designintention samt föreslå möjliga designändringar för att förbät-tra tillverkningsförmåga, analysera och granska 2D-ritningar och 3D-modeller, och utveckla ordernoteringar.
I det här arbetet har en prototyplösning tagits fram. Avsikten med denna är att överbrygga ett gap som har identifierats genom kartläggning mellan tillverkarens behov och funktionalitet av tillgänglig CAD / CAM-programvara och de identifierade förbättringsområdena från arbets-flödesutredningarna.
Den föreslagna lösningen, som presenterades i prototypen, har validerats tillsammans med tre av de studerade företagen. Vid sitt nuvarande tillstånd behövs ytterligare förbättringar och valideringar. Om den vidareutvecklas har den dock potentialen att skapa värde för hela tillverkningskedjan.
Preface
This thesis was completed as the final part of my master’s degree in Production Engineering and Management within the track Industrial IT systems at KTH Royal Institute of Technol-ogy in Stockholm, Sweden during the spring semester of 2019. It was carried out for Sandvik AB’s division Applied Manufacturing Technologies (AMT) at the Center of Digital Excellence (CODE) in Stockholm, Sweden.
I would like to express my deepest thanks and appreciations to my supervisor at Sandvik CODE, Dr. Vahid Kalhori, for great support and guidance throughout my thesis work and for challenging me in my way of thinking, and to my supervisor at KTH, Magnus Lundgren, for his encouragement and quick feedback about my work. I would also like to express my gratitude to my colleagues at Sandvik CODE for discussing various issues with me throughout this project.
Finally, I would like to show my greatest appreciation for the companies I had the opportunity to visit and all the respondents I got to interview. Thank you for sharing your valuable knowl-edge and experience with me, and for helping me better understand the challenges that you face in your everyday work. Without your help, this thesis would not have been possible.
Table of Contents
1 Introduction 1
1.1 Background . . . 1
1.2 Purpose & Aim . . . 1
1.3 Research Questions . . . 2 1.4 Delimitation . . . 2 1.5 Disposition . . . 2 2 Methodology 4 2.1 Research Design . . . 4 2.2 Sample Selection . . . 5
2.3 Data Collection Methods . . . 6
2.4 Data Analysis Methods . . . 9
3 Literature Review 10 3.1 Design and Process Planning Flows . . . 10
3.2 Integration Challenges and Approaches for Solutions . . . 11
3.3 Summary of Literature Review . . . 15
4 CAD Software Market 16 4.1 Market Insights . . . 16
4.2 Market Trends . . . 19
4.3 Summary of CAD Software Market . . . 24
5 Results & Analysis 25 5.1 Mapping of Workflows . . . 25
5.2 Identified Pains & Areas of Improvements . . . 34
5.3 Value Stream Mapping . . . 37
5.4 Value Delta . . . 39
5.5 Identified Market Gap . . . 41
5.6 Proposed Solution . . . 42
6 Discussion 48 6.1 The Research Questions . . . 48
6.2 Sustainability Aspects . . . 50
7 Conclusions 52
References 54
List of Figures
2.1 Research Phases . . . 4
3.1 Three KID Flows . . . 10
3.2 General CNC Machine Shop Workflow . . . 11
3.3 Comparison Between Generative Machining and CyberCut . . . 12
3.4 CAMA . . . 13
3.5 Tolerance vs. Cost . . . 13
3.6 Inter-Operational Issues . . . 14
4.1 Global CAD Market . . . 16
4.2 CAD Market, by Deployment . . . 17
4.3 The Top 15 CAD Packages . . . 20
5.1 Workflow Company A . . . 25 5.2 Workflow Company B . . . 27 5.3 Workflow Company C . . . 28 5.4 Workflow Company D . . . 30 5.5 Workflow Company E . . . 31 5.6 Workflow Company F . . . 32
5.7 Value Stream Mapping . . . 38
5.8 Value Delta . . . 40
5.9 Competitive Landscape . . . 42
5.10 CAD-CAM Solution . . . 45
List of Tables
2.1 Classification System . . . 5
2.2 Classification of Samples . . . 6
List of Abbreviations
CAD Computer Aided Design
CAM Computer Aided Manufacturing
CNC Computer Numerical Control
CMM Coordinate Measuring Machine
PMI Product and Manufacturing Information
GD&T Geometric Dimensioning and Tolerancing
DFM Design For Manufacturability
MBD Model-Based Definition
OEM Original Equipment Manufacturer
SME Small and Medium-sized Enterprises
APAC Asia-PACific
NAFTA North American Free Trade Agreement
AEC Architecture, Engineering and Construction
1 Introduction
1.1
Background
In today’s fast-paced market, it is more important than ever for manufacturing companies to stay competitive. This can be done by reducing time-to-market for new products, decreasing costs, and utilizing the production process in the most optimal way. Nevertheless, there are still many challenges with achieving this, one of which is the integration between design and process planning.
Traditionally, process planning problems have been solved with the experience and knowledge of experts. However, there are many issues related to this approach including its dependency of such expert; experts may be difficult to find and expertise takes time to develop. Process planning can also be error-prone as less experienced planners are likely to make more mis-takes, which results in high deviations in the manufactured products. Moreover, the design intent of the product, as well as process design, may not be properly communicated or un-derstood among all involved actors in the product realization process, which determines the conditions for achieving the right quality [1]. Consequently, this increases the lead time, the scrap production, and unnecessary rework, which overall results in increased manufacturing costs and less profitability.
Improved software integration in the product realization process is believed to solve many of these issues by enabling manufacturing enterprises to gain production-related benefits, such as reduced production cycle times, higher throughput, and increased product quality [1], [2]. One approach that could achieve a better-integrated design and process planning is through the support of cloud-enabled engineering.
1.2
Purpose & Aim
INTRODUCTION
proposition has been put forward with the intention of benefiting the entire value chain, up-stream and downup-stream, of the manufacturing industries.
Thus, the goals of this thesis were to:
I. Identify pain points and bottlenecks for the current workflow in design and process plan-ning through value stream mapping.
II. Identify available engineering tools for automated process feedback from CAM/CMM to CAD.
III. Identify existing platforms for cloud-based and on-premise CAD-CAM and analyze their impact on the value chain.
IV. Based on the above:
A Identify areas of improvements and propose solutions.
B Validate suggested solutions through conceptualization and contextualization through cloud-based services.
C Present a value proposition.
1.3
Research Questions
Based on the research objective stated above, the following research questions (RQ) could be formulated:
RQ1 What pains do manufacturing companies currently have in their workflow of design and process
planning?
RQ2 How can a cloud-based application increase efficiency and reduce waste in design and process
planning?
1.4
Delimitation
This thesis work has studied Swedish manufacturing companies in various sizes, various in-dustries, and with various operational models. The sample companies workflows were mapped, where only their pains related to the integration between CAD and CAM were identified. Thus, other pains that the manufacturing companies might have, which are not related to this area, were not covered.
1.5
Disposition
INTRODUCTION
2 Methodology
To understand and identify the industry needs, a combination of quantitative and qualitative methods were used. Six Swedish manufacturing companies, of different size, in different industries, and with different operational models, were selected as a sample group. The primary data was collected by visiting the sample companies for interviews and field observation as well as via phone and email. Literature reviews and other secondary data were collected throughout the research work. The collected data were analyzed using a thematic approach.
2.1
Research Design
The methodology for the research was a combination of quantitative and qualitative meth-ods, which is a recommended approach within business and management research [3]. It was conducted with concurrent cross-sectional design and followed a lean startup approach. The research process was divided into three phases (Figure 2.1).
FIGURE2.1: The three research phases.
METHODOLOGY
2.2
Sample Selection
The research samples were selected with a non-probability, typical case sampling approach with the intention of giving an illustrative representation of the research work. The selected sample group was Swedish manufacturing companies and were classified by company size, industry, and type of operational model. See Table 2.1 for the classification types and Table 2.2 for the classification of the selected sample companies.
TABLE2.1: Classification system.
Classification Options Description
Company size Small 1-15 employees
Medium 16-150 employees
Large 151-1,000 employees
Very Large >1,000 employees
Operational model Prototype Operating as independent shops, captive to larger
customers, or part of a prototyping unit of larger cor-porates. They have internal engineering competence providing CAD/CAM services to their customers. They are producing prototype parts and components on orders from their internal or external customers with varying batch size.
Repeating Independent companies, which are primarily
run-ning medium-to-large size jobs with repeating work orders. These are usually suppliers to companies within individual industry segments, e.g. automo-tive, aerospace, etc. who already have been certified to run part production of specific parts for the OEMs.
Non-repeating Independent companies, which are contractors to
one or several industry segments, running a large number of orders with different part number and various low-to-medium size volume. They are taking orders for production of parts from several different industry segments, e.g., both aerospace and automo-tive and machine tool manufacturer.
OEM Similar to a value-added reseller. It refers specifically
METHODOLOGY
TABLE2.2: Classification of sample organizations.
Org. Company size Type of industry Operational Model
A Large Automotive; Die & Mold Prototype
B Aerospace OEM
C Machine Tool Manufacture Prototype; Repeating
D Medium Aerospace Prototype; Repeating
E Small General Machining Prototype
F Machine Tool Manufacture; Die & Mold Non-repeating
2.3
Data Collection Methods
The primary data was collected by visiting the sample companies for interviews and field ob-servation as well as via phone and email. Literature review and other secondary data were collected throughout the research work. The different methods of data collection are presented below.
2.3.1 Literature Review
Secondary data can provide additional knowledge, interpretations, or conclusions to the re-search work [4]. The secondary data in this rere-search work was collected to mainly support the literature review and the CAD software market analysis.
Scientific articles and books were used to support the literature review, which was collected throughout the research work. The main sources of information were KTH’s library’s database Primo, Google Scholar, and Google searches.
CAD market reports, surveys from CAD organizations, CAD magazines, and blog posts were used to support the CAD software market analysis. Here, the main source of information was Google searches.
2.3.2 Interviews & Observations
METHODOLOGY
TABLE2.3: An overview of the respondents and the date of the interviews.
Resp. Role Experience Org. Date
R01 Production Manager Has about 35 years of experience in the
industry. Went directly from 2 years of upper secondary school to work. Has experience as an operator, production technician, and manager.
D 2019-03-18
R02 Quality Manager Has about 30 years of experience in the
industry. Went directly from 4 years of upper secondary school to work as a manual machine operator.
R03 CAM programmer
& Machine Operator
Has about 25 years of experience in the industry. Went directly from school to work as a machine operator. Has used CAM software for about 20 years.
R04 Managing Director Has an MSc in Mechanical Engineering.
Has worked as a calculation engineer for several years. Has many years of experience with CAD. Has 1.5 years of experience with CAM.
E 2019-03-20
R05 Managing Director Has an MSc in Mechanical Engineering.
Has worked in various industries with technical sales, process engineering, as a managing director, and as a consul-tant.
F 2019-03-20
R06 CAM programmer
& Machine Operator
Has worked as a machine operator for 5 years. Has a couple of years of expe-rience with CAM and CAD.
R07 CAM programmer
& Machine Operator
Went directly from industrial upper secondary school to work as a machine operator. Is a beginner with CAM and CAD.
R08 Managing Director Has about 20 years of experience in the
industry. Went directly from industrial upper secondary school to work. Has worked as a machine operator, produc-tion technician, and process planner.
A 2019-03-25
METHODOLOGY
Table 2.3 – continued from previous page
Resp. Role Experience Org. Date
R09 Process Planner Has several years of experience from
the industry, working as a machine
op-erator. Went directly from industrial
upper secondary school to this organi-zation.
R10 CAM programmer
& Machine Operator
Has about 15 years of experience as an operator. Went directly to working from industrial high school. Has 2 years of experience with CAM software.
R11 Product
Specifica-tions and Verifica-tion Manager
Went from industrial high school di-rectly to working at his current organi-zation. Has worked as a machine oper-ator, technician, and process planner.
R12 Inspection Manager Has several years of experience in
the industry. Has worked as a
ma-chine operator and production techni-cian, among other roles. Went directly from high school to work.
B 2019-03-26
R13 Quality Manager Has several years of experience in the
industry. Has worked as a machine op-erator before this position.
R14 Mechanical Design
Engineer
Has about 5 years of experience in the industry. Has studied a BSc & MSc in Mechanical Engineering. Has worked as a production technician, quality con-troller, and with R&D.
C 2019-03-27
R15 CAM programmer Went directly from industrial high
school. Has worked in the industry
since 2001, first as an operator and later as a process planner. Has 9 years of ex-perience with CAM software.
R16 Process Technician Has 35 years of experience in the
indus-try. Started on the shop floor and has experience from several roles.
R17 Quality Control Went directly from upper secondary
METHODOLOGY
A total of 17 respondents were interviewed between mid-March to the end of March of 2019. The interviews had a semi-structured format to allow for a complete exploration of the research questions within selected key themes and questions, which is suitable for an exploratory and explanatory research approach [5]. The interviews were conducted, when it was possible, at the work desk of the interviewees. Despite often very disturbing environments with loud noises from the machines and other distractions, these location settings were preferred due to the na-ture of the research work. The interviews were capna-tured through audio recording and note taking. During the interviews, observational notes were also taken. After this, the interviews were summarized and relevant data samples were transcribed. The interviews were not tran-scribed as a whole as a result of the distracting environments and for time efficiency. The interviews were conducted in both Swedish and English, where the Swedish quotes have been translated.
2.3.3 Survey
In order to collect more quantitative data, the sample companies, excluding Company B, were asked to answer additional questions about their workflow. The data was collected through email and phone. These phone interviews were captured through note taking.
2.3.4 Validation
The validation sessions were conducted via video calls with three of the sample companies (Company C, E, and F), where the received feedback was collected through note taking.
2.4
Data Analysis Methods
3 Literature Review
Traditionally, design and process planning can be described with three KID (knowledge, information, and data) flows. In these flows, there are several integrational and inter-operational challenges between various software used in the manufacturing value chain. Numerous studies have been done with the aim of closing the knowledge gap between design and process planning and making the integration of different CAx software more seamless and better-integrated, but with various success.
3.1
Design and Process Planning Flows
Interactions in design and process planning can be described in terms of three knowledge, in-formation, and data (KID) flows [7], as illustrated in Figure 3.1. The first flow happens when the designer provides design and engineering data to the manufacturer. The second flow is the request for design changes or corrections from the manufacturer to the designer, which the designer can either accept or reject. This flow changes throughout the development of the product, with more informal communication in the early stages and more formal communi-cation in the final stages of product development. The third flow is the post-analysis results, e.g. a quality assurance report, from the manufacturer to the design team, however, there is normally no structured way of capturing and reusing this information in most organizations.
LITERATURE REVIEW
The CAD-CAM supplier BobCAD describes the information flow of many of its customers as described by Figure 3.2. According to them, most CNC machine shops start off with a 3D model of the part, either directly provided by the customer or created by themselves from a 2D drawing provided by the customer. The 3D model file is then imported into a CAM system or an integrated CAD/CAM system. Here, tool paths, setups, and operations are decided before final machine simulations and G-code can be generated. The G-code is then sent to the CNC machine where the part can be machined. After quality inspections, the final product can be delivered to the customer [8].
FIGURE 3.2: A holistic view of a workflow for a general CNC machine shop (modified) [8].
3.2
Integration Challenges and Approaches for Solutions
There are several challenges in the integration of design and process planning. One struggle is to design parts, components, and products that are cost-effective, or maybe even possible, to manufacture. Design for Manufacturability (DFM) is an approach that tries to solve this issue and consists of set guidelines and rules for the designer to follow in the early stages of product development when design issues are the least expensive to fix. The DFM method has shown to be beneficial if applied correctly. According to DFMPro, a CAD-integrated design for manufacturing software, using such design practices efficiently can reduce rework with about 20%, production cost with about 6%, and time-to-market with about 10% for manufacturing companies [9].
LITERATURE REVIEW
of the identified research issues was deficiencies in software tools supporting the design pro-cesses. This was confirmed by Zwolinski et al. [12] in their work about how virtual reality technologies can support designers in transforming the definition of product functionalities to the design of its structure. Moreover, Lundgren et al. [1] point out that the creative part of process design has yet to be fully researched and emphasize the importance of being able to communicate the design intent of both process and product design efficiently.
To close this research gap, Ahn et al. presented their research project CyberCut [13], a web-based Java CAD/CAM system which used DFM rules and guidelines in the design phase to make manufacturing constraints obvious for designers in addition to making process planning more seamless and automated by, for example, automatically generating tool paths for a 3-axis CNC milling machine from the incoming design (see Figure 3.3). Molcho et al. [7] presented a holistic approach and supporting software tool referred to as Computer Aided Manufactura-bility Analysis (CAMA), where the motivations, as well as the required components for such a system, were described. The proposed system was embedded in CAD software, where the analysis of design is performed in accordance with several DFX guidelines and presented in a digital report as well as graphically. The authors claimed that CAMA makes “know-how” available for designers at an early stage of the design process by enabling quick analysis of a design, which is expected to close the gap between designers and manufacturers (see Figure 3.4). In the PLM/CAD-CAPP field, Denkena et al. [14] presented a developed ontology for the process planner environment as well as models for more efficient design-making and knowl-edge management. These models were based on integrated business and technical aspects with the intention of improving available PLM/CAD-CAPP software solutions.
LITERATURE REVIEW
FIGURE3.4: a) CAMA components. b) CAMA analyzer [7].
Geometric tolerances are part of the DFM guidelines because they heavily contribute to the cost of machined components. The Department of Mechanical Engineering at The Ohio State University illustrates the relationship between tolerances and production cost as seen in Figure 3.5 [15]. Needless to say, critically analyzing set tolerances to remove any unnecessary ones could greatly reduce the manufacturing cost of a part.
FIGURE3.5: An illustration of how tolerance and production cost are correlated [15].
LITERATURE REVIEW
formats, and so on. In a survey among 251 executive officers, the biggest struggle in the inter-operation of software systems in the CAD-CAM-NC chain was that they had “different CAD formats used by customer and supplier” (Figure 3.6) [16].
FIGURE 3.6: The main inter-operational struggles of software systems in the CAD-CAM-NC chain. [16]
One of the most successful solutions and a key enabler for neutral data exchange and integra-tion between various CAx systems, such as CAD, CAM, and CAE, is believed to be STEP (STan-dard for the Exchange of Product model data) [17], [18]. Formally known as ISO 10303, the in-ternational standard for exchanging product manufacturing information, which uses domain-specific application protocols (AP) to incorporate the requirements for the product data models in their respective domains.
LITERATURE REVIEW
Apart from its neutrality and compatibility with various CAx system, STEP files can also be used to protect sensitive company data and intellectual property. Nevertheless, there are limi-tations with how available software systems implement the standard.
3.3
Summary of Literature Review
4 CAD Software Market
In 2017, the global CAD software market was valued to 8.325 billion USD and was forecasted to ob-serve a compound annual growth rate (CAGR) of 6.8% over the years 2018 to 2026, reaching over 14 billion USD by 2026. With regards to deployment, the cloud-based CAD was 39% and is expected to be a contributing factor for the entire market’s growth. In general, trends in the CAD market are strongly focused on higher accessibility and flexibility, which demands better collaboration, integration, and mobility of CAD software.
4.1
Market Insights
In 2017, the global CAD software market was valued to 8.325 billion USD [22] and was fore-casted to observe a compound annual growth rate (CAGR) of 6.8% over the years 2018 to 2026, reaching over 14 billion USD by 2026 [23], [24] (Figure 4.1).
FIGURE4.1: The global CAD software market and its expected growth.
CAD SOFTWARE MARKET
CAD software market in the coming years and contribute greatly to the strong growth potential [25].
4.1.1 Deployment
The CAD software market can be categorized into two types of deployment: on-premise and cloud-based. The on-premise CAD software is directly installed and runs on the end-users computers, whereas cloud-based CAD software is hosted in the cloud and can be accessed anywhere at any time.
In 2014, on-premise CAD solutions were the dominating deployment of the two with close to 89% of the revenue share [26]. However, as more companies have adapted to cloud-based solutions, the cloud-based CAD market has grown significantly and had, in 2017, close to 39% of the market share [22] (Figure 4.2). Moreover, it is expected to persist this growth during the coming years [25], [26]. This is especially true in the mid-market segment, where cloud-based CAD software is growing significantly due to the lowered investment costs. These users purchase CAD software packages but they are more price sensitive than the high-end market segment [27].
FIGURE4.2: In 2017, close to 39% of the market share was cloud-based CAD. [22]
CAD SOFTWARE MARKET
4.1.2 Application
The key applications for CAD software are in the manufacturing, automotive, architecture, engineering and construction (AEC), healthcare, and aerospace & defense sectors [22], [26]. The manufacturing sector is predicted to continue to grow during the coming years as a result of emerging innovations, which will call for better-integrated CAD/CAM software solutions as manufacturers are continuously looking for better ways to improve product quality, reduce de-velopment time, and increase production efficiency [26]. An example of such innovation is the computer-integrated manufacturing (CIM) approach, where computers control and regulate the entire manufacturing process. Here, CAD is a key enabler as it makes product develop-ment much more flexible, efficient, and cost-effective [23]. According to Grand review’s report, the growth in the manufacturing industry and the growth in CAD software is believed to be highly correlated [26]. CAD software facilitates CNC systems to be better integrated and easier to use, which has resulted in a steady market growth since the introduction of CAD software [23]. Moreover, in the manufacturing industry as well as other industries, digital 3D modeling has proved to be an effective way of communicating between different stakeholders. This has contributed and is expected to continue to contribute to the CAD software market’s growth [26].
In the aerospace & defense sector, CAD software allows for better precision in designing to-gether with simulation software [22]. In the automotive sector, 3D CAD and rapid prototyping have been important factors for reduced time-to-market [26]. In the healthcare sector, where the dentistry industry is especially driving the growth [22], CAD can, for example, be used for creating personalized medical products and devices [26] or to excerpt data from computer tomography scanners to generate 3D models [23]. In the AEC sector, 2D CAD has been used for drafting and sketching of constructions, but 3D CAD is increasing in importance. This wide range of applications is believed to ensure the continuous growth of the entire CAD market for the distant future.
4.1.3 Regional
In 2017, North America was dominating the CAD software market with more than one-third of the global revenue share [23], [24], [26]. The usage of CAD software in this region is widespread among a variety of applications, sectors, and end-users. It is especially large in the manufac-turing, automotive, and aerospace & defense sectors in this region, where 3D modeling and rapid prototyping are essential to reduce the time-to-market for new products. Consequently, the market is expected to stay resilient over the forecasted period [26].
CAD SOFTWARE MARKET
the development of mobile CAD [22], [25] as a result of its high maturity of mobile devices [28] and demand for sketching and drawing apps for product design [22]. In terms of the market value of cloud-based CAD software, the APAC region will also have the largest growth [25], [26].
4.1.4 Usage levels and CNC users
The CAD market can be divided into three main segments based on usage levels: professional, intermediate, and beginner. Professional users tend to use the most expensive, complex prod-ucts offerings; intermediate users tend to pay for the software but not as much as professionals; beginners will likely use free software packages. In 2017, the intermediate level was dominat-ing the market with over 45% of the share [22].
Among CNC users, a fourth group of users can also be identified, namely those who use the built-in CAD module in their CAM packages. In a survey with just under 400 responses done in 2018 by CNCCookbook, a CNC software provider and blog with about half of the readers based in the US [27], 29% of the respondents said that they used free CAD, which also included educational and pirated software. Fusion 360, an integrated CAD/CAM/CAE software, which is available as both a free and a paid package, was the biggest player both among free users and paid users. See Figure 4.3 for the top 15 CAD packages used in production, according to the CNCCookbooks’s surveys from 2015 to 2018 [27].
4.2
Market Trends
The CAD software market is dynamic and has several promising trends to contribute to its growth in the coming years. New, emerging technologies are expected to have the biggest growth potential [25], making CAD software overall more accessible, flexible, and user-friendly [23]. It can be observed that several of the trends that will contribute to the growth of the CAD software market are dependent on cloud technology.
CAD SOFTWARE MARKET
FIGURE 4.3: The top 15 CAD packages used in production, according to CNC-Cookbooks’ CAD surveys [27].
The key trends in the CAD software market and related technologies that have been identified and will be further described are:
• Mobile CAD • Cloud-based CAD • Collaborative design
• Subscription-based solutions & remote training • Augmented Reality (AR) & Virtual Reality (VR) • Simulations & generative design
CAD SOFTWARE MARKET
4.2.1 Mobile CAD
Today, mobile access is almost required by customers in nearly every technology [23]. Mobile CAD has a strong growth potential [25] and the market for it is growing across all sectors, both together with desktop CAD and on its own. Market analysts are not expecting it to replace desktop solutions, but rather replace on-site activities that often are performed with paper and pen [28]. For instance, in the Business Advantage report, it was stated that about half of the respondents were aware of the trend, but merely a fifth of them was using it currently. Among the current and future users, viewing and annotating CAD files were/expected to be the main usage for mobile CAD. Only 35% were expected to create CAD files using mobile devices [25]. As a result of this and to the continuous development of mobile hardware devices, it is likely that CAD software with deeper 3D support will evolve. The growing demand for higher mo-bility will highly affect the workflow of product development and is expected to drive better collaboration and integration between users. Furthermore, this will likely increase the produc-tivity of companies as knowledge is better shared and production cycles can be shortened. Due to this, synchronizing files and handling version control efficiently across all devices and users will become even more essential than as of today [28]. The demand for mobility of portable devices is high for both large and small companies regardless of the end-user [22], however, it is expected that companies in the APAC region will drive the development of mobile CAD [25].
4.2.2 Cloud-based CAD
Over the next 5 to 10 years, it is believed that the majority of all work will move online. This, together with increasing trends of higher mobility, accessibility, and a progression of cloud storages, could be contributing factors to the growing demand for cloud-based CAD solutions [23], [28]. Cloud-based CAD solutions are expected to grow especially among SMEs and in-termediate users thanks to possibilities similar to most cloud-based solutions, namely its high scalability and high flexibility. In addition, cloud-based solutions enable innovative business models and new target groups by lowering the threshold of expensive investments and com-plicated maintenance that most off-premise solutions require [22], [26].
CAD SOFTWARE MARKET
The main drivers for cloud-based CAD are higher mobility, ease of updating software, and expected cost reductions [23], [25]. Among the respondents of the Business Advantage report, the current or expected use cases of cloud-based CAD were the collaboration of CAD files, review of CAD design files, and storage and backup of CAD files. About a third of the users were expecting to use CAD hosted in the cloud when designing. Moreover, 20% of the users were expecting to use cloud-based CAD as their primary CAD program [25]. It is believed that as cloud-based CAD becomes higher adopted, developments in parametric modeling and data management will make big impacts in the product development process [28].
Software upgrading is a major issue for on-premise CAD users and can result in lost time and costs [29]. For instance, in a survey conducted by Onshape [30], 78% of the respondents said that managing infrastructure and upgrades were one of their main pain points when using on-premise CAD. Similar results are presented in the Business Analysis report [25]. Here, 24% of the respondents had productivity loss while adjusting to an upgraded version of their CAD software. The length of the productivity loss was, on average, 1-2 weeks and only 23% of those who upgraded achieved productivity advantages afterward. In addition, many companies skip years of upgrades on purpose to avoid the risks and software bugs from upgrading. With cloud-based CAD, software upgrades are never needed [29].
4.2.3 Collaborative design
Similar to mobile CAD and cloud-based CAD, collaborative design is a growing trend in the CAD software market. It benefits the users by speeding up design time, reducing errors, im-proving the designs, saving on costs, and enabling designers to be located in multiple locations. The usage of collaborative design has increased the most when compared to mobile CAD and cloud-based CAD [25].
4.2.4 Subscription-based solutions & remote training
Cloud-based solutions have enabled new business models to be formed. Recently, a shift from license-based to subscription-based CAD software has been observed [22]. Normally, licens-ing means ownership of the software as a one-time investment, whereas subscription-based models offer the service at a lower cost, likely making the entry barrier lower for end-users. Consequently, many users that are more sensitive to pricing, normally hobbyists and SMEs, are able to access higher-end CAD software at a lower price. Today, license-based CAD is dom-inating the global CAD market. According to the Business Analyst report, 70% use a perpetual license with or without annual support and/or maintenance and only 37% use an annual sub-scription as their primary CAD software. Only 4% use a subsub-scription license on a shorter term than a year [25].
CAD SOFTWARE MARKET
into actual productivity. That way, new applications can be successfully implemented among the users [28].
4.2.5 Augmented Reality (AR) & Virtual Reality (VR)
Augmented Reality is a technology that is believed to become essential for product developers in the coming years [28] and has strong future growth potential, especially among larger firms. The technology can augment CAD models with direct or indirect real-world environments [25], which means that 3D models can be placed directly onto our current reality and experienced throughout the entire product development phase. This kind of reality modeling is expected to grow in importance and has shown to be a clever way to immersive customer experience [28]. Although expected to be used in all sectors, AR is believed to have the biggest impact in the AEC industry [25].
Similar to AR, Virtual Reality (VR) is expected to have strong growth potential in the CAD market in the coming years, which is likely to be driven by larger companies. The technology has similarities with AR, but instead of placing 3D models in a real-world environment it is generated in a simulated environment. However, the use case between AR and VR for product development is believed to be similar, where both technologies are predicted to mainly be used for communicating design intents to clients [25].
4.2.6 Simulations & generative design
In several sectors, there is a growing need for precise simulations and real-time renderings [22]. This has encouraged the development of simulation software, such as ANSYS, to become better and quicker. For instance, one simulation tool completely removed the need for meshing, which is one of the most tedious tasks when performing a simulation [28]. Such advancements also help the designers by speeding up the design process. Furthermore, generative design tools are also advancing, which will likely optimize simulation-driven designs. Daniel Graham, the director at Fusion 360, believes that this new technology could completely change the way we design new products [28].
4.2.7 Internet of Things (IoT)
CAD SOFTWARE MARKET
4.2.8 2D drafting & 3D modeling
For small and medium-sized companies, 2D drafting is still viewed as the most important core business function in CAD [25]. Nevertheless, a major trend among smaller business is to shift from 2D drawings/drafting to 3D modeling. This trend is especially true for the AEC sector, where designs are expected to be transformed from 2D drawings to 3D models or created di-rectly as 3D models [28]. This transition will help designers analyze and simulate their design as well as digitally prototype the designs before production [22].
4.2.9 CAD/CAM integration
Another trend is that the integration between manufacturing and CAD is becoming tighter. Traditional product development methods are replaced by more agile approaches, which an-ticipates that data management will become crucial for manufacturing companies [28].
4.3
Summary of CAD Software Market
5 Results & Analysis
The sample companies workflows were mapped and three major areas of improvement could be identified in their manufacturing value chain. A value stream mapping was outlined from the request for quotation to confirmation of an order as well as its value delta. An identified gap in the market, as well as a proposed solution, a mock-up prototype, and validation feedback from this mock-up, were put forward. After each section, a short analysis of the results can be found.
5.1
Mapping of Workflows
5.1.1 Company A
FIGURE5.1: Mapping of the current workflow for Company A.
RESULTS & ANALYSIS
either in its native or neutral file format without any PMI in the model. The 2D drawing is always printed and used as a reference document for the part’s PMI.
First, a process planner reviews the order and if changes are needed, he/she requests it to the customer via a helpdesk ticketing system, email, or phone. Afterward, he/she plans the process for the part as well as estimates the time and cost it would take to for it to be manufactured. The process planners estimate that they spend about 50% of their time discussing the designs with the customers and 50% of their time doing process planning. Thus, with three full-time process planners and 40 hour work weeks, this results in 60 working hours per week spent on discussions with the customers regarding the design. The discussions normally consist of understanding the design intent and clarifying measurements and tolerances on the drawings. If changes are needed, the updated files are received within an hour (10%), during the same day (20%), during the next day (45%), within a week (20%), or more than a week (5%). Moreover, the process planners find it difficult to plan the process of the part and to estimate the time and cost correctly. Company A does not use any CAPP system.
After that, a process plan and the printed 2D drawing are given to the machine operator, who downloads and imports the 3D model from the helpdesk ticketing system into the CAM sys-tem. While programming, tolerances are highlighted on the drawing to keep track of which ones that have been taken into account for in the CAM program. The received 3D models are made in basic size, but to achieve the desired tool paths at the nominal size, new support lines in the CAM system has to be created. When the machine operator is satisfied with the result, he/she simulates the part in the CAM system, post-processes the program, and uploads the created NC code to the network server. After that, the machine operator manufactures the part and performs a quality inspection with manual tools. Company A does not use a CMM ma-chine, however, they do receive orders with GD&T requirements. In those cases, it is assumed that the part is manufactured correctly. Company A estimates that approximately 3% of the manufactured parts are scrapped due to parts not meeting the set tolerance requirements.
5.1.2 Company B
The workflow for Company B can be seen in Figure 5.2. Company B has an MBD driven workflow. First, the part is designed by in-house mechanical design engineers and uploaded to the company’s PDM system. The 3D model is created in nominal size, containing all of the part’s PMI, and is available in its native file format. The process planners transform the 3D file format into a lightweight model using a script and uploads it to a web client where all the manufacturing data is located. The CAM program is created from the 3D model and is validated with a CNC simulation and validation system. However, the process feedback from this validation is not collected.
RESULTS & ANALYSIS
FIGURE5.2: Mapping of the current workflow for Company B.
and uploaded to the PDM system. However, the data from the reports are not collected in any systematic way or looped back into the process.
5.1.3 Company C
The workflow for Company C can be seen in Figure 5.3. Company C has the entire CAD-CAM-CNC-CMM value chain in-house, where mechanical design engineers receive customer orders via an order form where features, PMI, requirements, and other details can be specified. Based on this, the mechanical design engineer models the part in the CAD system. At the time of the data collection, a process for automatically creating models in CAD based on the order’s specifications was under development. Once the 3D model and the 2D drawings are completed, they are uploaded to the company’s PDM system. The 3D model is uploaded in its native file format, but with the possibility to download it in its neutral file format directly from the PDM system.
After that, the process planners work will begin. Currently, Company C has two main work-flows for manufacturing, where one is focused on prototypes and one is focused on medium to large-sized series.
RESULTS & ANALYSIS
and the end customer’s priority. Next, the process planner uploads the 3D model’s native file format into the CAD system, transforms the model from its basic size to its nominal size with a built-in automatic tool in the CAD system, and exports the 3D model as a STEP file. After that, the process planner imports the file into the CAM system and starts programming. If the
RESULTS & ANALYSIS
process planner needs to do changes in the model to create the desired tool paths, the CAD module in the CAM system is used. For very complex parts, the original CAD system might be used. After the operations are completed, the part is simulated and later post-processed. Then, the created G-Code often needs a few manual changes, due to issues in the system, before it can be transferred to the CNC machines on a flash drive. After the part has been manufactured by the machine operator, quality inspections are done manually. They estimate that 1-2 minutes are spent on quality inspections for a part that takes 10 minutes to machine. The scrap rate for this manufacturing workflow is 10-12%.
In the second workflow, the process planner receives the 2D drawings via email from the me-chanical design engineer to review them. The process planner estimates that he/she spends approximately 8 working hours every week only on reviews. Moreover, the company has meetings twice a week for one hour, where process planners, operators, and quality controllers meet to mainly discuss tolerances and manufacturability of the incoming orders. About five employees normally attend these meetings, thus resulting in 10 working hours per week spent on reviews. The process planner experiences that they are answering similar questions repeat-edly, especially about production cost with regards to tolerances. After the 2D drawings have been agreed upon, the process planner downloads the 3D model in its neutral file format from the PDM system, imports it into one CAD system to create it in a 2D view, exports and im-ports it to another CAD system to create the 2D drawings for the process plans. After that, the machine operators receive the process plans and programs the NC code with the support of a system called WindFlex. After the parts have been machined, quality inspections are done manually at the CNC machines by the machine operator, who follows a checklist, and later in the CMM machine by the quality controllers, where a quality report is generated. The aver-age CMM programming time is 2-3 hours and there is no process feedback from the quality controls. The scrap rate for this workflow is 3%.
5.1.4 Company D
The workflow for Company D can be seen in Figure 5.4. Company D receives its customer orders via email, in person or via a network server. The order normally contains a 2D drawing, a 3D model in its neutral file format, and other specifications of the part. Depending on how early Company D gets involved in the customer’s design phase, they sometimes only start off with a napkin drawing from the customer. Company D estimates that they spend roughly 20 working hours per week discussing customer’s designs, but claims that the earlier they get involved, the better.
RESULTS & ANALYSIS
FIGURE5.4: Mapping of the current workflow for Company D.
programming time as well as increases the risk for rework. In the rare cases of models in its nominal size, errors happen anyway due to it not being clearly communicated. However, different mechanical design engineers have different ways of modeling their parts and with recurring customers, the process planner often learns the mechanical design engineer’s intent with the design.
After simulating the program in the CAM system, the program is post-processed and the NC code is uploaded to a server. The person preparing the process planning is also the same person machining the part, and after the part has been machined, they also inspect the part with manual inspection tools. For more complex parts, parts with GD&T, or if the customer has required it, CMM are also used. The CMM programming of a part takes between 1 to 16 hours, depending on the part’s complexity and the required documentation. The average time is estimated to approximate 7 hours. Company D does not track its scrap rate.
5.1.5 Company E
RESULTS & ANALYSIS
FIGURE5.5: Mapping of the current workflow for Company E.
Afterward, the process planner reviews the received order within 24 hours to ensure its man-ufacturability. The web application is not yet capable of analyzing the design’s manufactura-bility when giving a quote. If changes in the model are necessary, the process planner either requests this from the customer or changes the part by him/herself and confirms the changes with the customer through email or phone. Company E estimates that they spend about 5% of part’s total manufacturing time discussing the geometry with the customer.
After the design has been agreed upon, the part is programmed in CAM, NC code is uploaded to a network disc, and the part is machined. If reworks are necessary, changes are made directly in the CAM program and post-processed to the machine until the desired result is achieved. Next, quality assurance is done on the part to fulfill all of the tolerances on the 2D drawing. Manual measuring tools are used for a majority of the parts, but for stricter tolerances, prob-ing in the CNC machines are done. Company E estimates that about 5% of the part’s total manufacturing time is spent on quality assurance.
5.1.6 Company F
RESULTS & ANALYSIS
as possible due to it being time-consuming. The reviewing of drawings are done relatively quickly, but if changes in the design are necessary, the updated model is received from the customer within one week on average.
FIGURE5.6: Mapping of the current workflow for Company F.
RESULTS & ANALYSIS
inspections are done manually and quality reports are only generated if requested by the cus-tomer. At the time of the data collection, Company F had just started tracking the scrap rate and its cause and, consequently, their numbers were believed to be inaccurate. However, the current scrap rate was 0.4%, where 75% of that was manufacturing faults.
5.1.7 Analysis of Mapped Workflows
Based on the workflow mapping of the sample companies, similarities and differences could be seen in regards to the size of the company, the different industry segments, and the operational models.
In the small- and medium-sized manufacturing companies (represented by Company A, D, E & F), the same person was responsible for the CAM programming, operating the machines, and quality assurance, however, in the large companies (Company B & C), these tasks were done by separate employees and even separate departments. When it comes to process planning, the small- and medium-sized companies did not use any CAPP systems to support this task, in contrast to the large companies, but instead was done in the process planner’s head based on experience. Among the small- to medium-sized companies, Company A was the only firm that formally documented the process plan and that had specifically assigned employees for process planning activities.
The large manufacturing companies had everything in-house, from CAD to CMM, which al-lows them to use a PDM system, share models in their native file format as well as connect features and data throughout the value chain, for example, with a model-based driven work-flow, see Company B. The small- to medium-sized companies, on the other hand, received models in a neutral file format (STEP), often via email. Here, Company A is an exception. Being a captive SME to a large company, they received models in their native file format and had a more-or-less guaranteed supply of new work orders. Apart from that, they had more similarities with the small and medium-sized companies.
Looking at the different industry segments, a couple of conclusions can be drawn. For the two companies in the aerospace industry (Company B & D), the components had strict and many tolerances as well as high requirements of traceability and documentation of the quality inspections. This explains the need for CMM machines. In contrast, the small and medium-sized manufacturing companies, that were not in the aerospace industry, did not use a CMM machine due to it being unnecessary, time-consuming, and/or expensive investment. Instead, they relied on a correct setup and machine probing in the CNC machines, even for parts with GD&T.
RESULTS & ANALYSIS
workflows with two different operational models, for instance, where the prototype work-flow’s scrap rate was 10-12% and the other workflow was down to 3%.
5.2
Identified Pains & Areas of Improvements
Based on the sample companies’ workflows above, the following pains and areas of improve-ments could be identified, presented in a thematic approach.
5.2.1 Pain #1) Understanding and discussing the design intent, manufacturability, and tolerances are time-consuming, repetitive, and difficult to communicate efficiently
After receiving an order, the drawing and/or model was reviewed and for many of the sample companies, several aspects had to be clarified with the customer before manufacturing could be started. First of all, the orders were in many cases incomplete, where the manufacturer had to spend time collecting more data before they could start. A Quality Manager described the lack of manufacturing information received from the end-customer:
“Can you please start? With what? What do you want? What color do you want? What material do you want? What are the holes for?”
- R02, Quality Manager
After collecting all the data, the sample companies spent time reviewing the drawings and/or models. A couple of the companies often received 2D drawings with too strict and too many tolerances than necessary as well as ambiguous tolerances. A few of the companies stated that they spend several hours per week reviewing and discussing the part’s set tolerances. A Process Technician described how they work with reviews:
“And the thing is that this is not a single-man race from my side; it isn’t just me that review them. Earlier I did it all by myself, but now I want to have a broad anchoring so I have all of my colleagues with me. So we have two days per week where I get to have a meeting where we walk through all new
articles, together.”
- R16, Process Technician
A CAM programmer answered a question about what ‘mistakes’ they often find on the files that they receive from their customers:
“Everything from tolerances to corners that are impossible to manufacture. Often, it is on the 2D drawing actually. Tolerances can be completely unnecessary. [...] Then you almost have to check with
the design engineer - is it important or not?”
RESULTS & ANALYSIS
When the companies reviewed the tolerances on the 2D drawings and/or 3D model, they wanted to be certain that they were able to deliver and validate the requested quality by hav-ing the necessary machine capabilities, tools, material, and quality inspection equipment that was needed. These sorts of reviews can be very depended on expert knowledge. A Managing Director described their workflow as follows:
“I will often prepare raw material, cost, etc. and then I give it to [the production manager] and we look at it together and [the production manager] says ‘This will take this long time, this will be the approximate cycle time’. [The production manager] calculates the amounts of tools needed, which
machines it can be done in, and so.”
- R05, Managing Director
Another Managing Director described their workflow:
“I take a look at the drawing to see, okay, what kind of different stuff I need. [...] And I just have that in the back of my mind when I’m doing the steps.”
- R04, Managing Director
When the companies discussed design changes with their customers, some of the requests were only to ease the manufacturing process. One CAM programmer & Machine Operator described the flow of unnecessary features in the models:
“Then, you have to call and check if it should be radii 8 for the whole part. And in most cases, it almost never has to be.”
- R02, CAM programmer & Machine Operator
Moreover, understanding and communicating the design intent and design changes may be difficult, depending on the complexity of the part and the change. These sorts of discussions with the customer were in many cases done over the phone or via email. When possible, it was also done face-to-face, but depending on the location of the company and their customers, this was not always possible. A major issue with communication design intent and tolerances over the phone or via email is that there is, apart from that it is difficult to communicate without the aid of visual tools, a high risk of misunderstandings. Thus, it can increase the risk for errors and the need for rework. One company, Company E, solved this issue by using Skype to share the screen to show the intended design changes in CAD when it was difficult to communicate over the phone or via email.
5.2.2 Pain #2) Keeping track of GD&T when CAM programming is difficult and error-prone
RESULTS & ANALYSIS
application of the part and the industry. When doing the CAM programming, the process plan-ner needed to remember the tolerances on the 2D drawing. A CAM programmer & Machine operator described it as follows:
“... and now I know that these two are on the same level [points on the computer screen]” “How do you know?” “Because I remember”
- R03, CAM programmer & Machine operator
The 3D models that the manufacturing companies received, were done in basic size, likely as a result of design practices based on form, fit, and function. In manufacturing, the nominal size is often desired. However, customized tolerances, which are specified on the 2D drawing, may be asymmetrical. This means that the model’s basic size is not equal to its nominal size. To solve this issue, the process planners adjusted for it by overcompensating it in the CAM software, adding extra support lines, and/or adjusting it manually in the CNC machine. The nominal value was calculated by hand and tolerances were tracked by marking it on the 2D drawings with a pen, or simply by remembering it. A CAM programmer & Machine Operator, who used a highlighter to help remember tolerance, described his/her workflow like this:
“I always use the highlighter. So then I mark the tolerance that I know that I have to re-draw. Then, when I sit down to re-draw it, I check off the ones that I have taken care of.”
- R10, CAM programmer & Machine operator
Dealing with asymmetrical tolerances when CAM programming was an annoying step for many of the CAM programmers. One CAM programmer & Machine Operator described his/her frustration as follows:
“Then it would also be nice to have them [the 3D models] modeled in the middle of the tolerance as well. If you have minus 400, you want to program it to 200 and then you have to do it manually. Otherwise, it would just be to click on a line, knowing that it’s correctly programmed. [...] Otherwise, you have to
transfer the information that is here [2D drawing] in there [CAM program]. Then when the next guy comes and opens it, then he doesn’t know how it’s programmed. So he runs it, and then the measurements are incorrect. Sometimes there are a lot of operations, then you need to find: where is the
middle?”
- R06, CAM programmer & Machine operator
Apart from being frustrating for the process planners, it may also heavily increase the pro-duction cost and time. The more tolerances to keep track of, the more likely it is for errors to happen. Thus, the scrap rate will likely increase. As one of the CAM programmers & Machine Operators expressed it when asked how much time he/she spends on this:
“It’s hard to say, but it is a frustrating aspect. And of course, it causes new problems as well. If it’s wrongly modeled, for example, and no one notices it.”
- R06, CAM programmer & Machine operator
RESULTS & ANALYSIS
5.2.3 Pain #3) Quality Inspections are done with CMM machine, which is very time-consuming, or without CMM machines, where GD&T cannot be vali-dated.
Quality inspections at the sample companies were done in two main ways: manual inspections at the CNC machine and/or CMM programming. For the companies without CMM machines, inspections were done with manual tools or with the help of the probing function in the CNC machine for stricter tolerances.
For GD&T on parts, which cannot be validated without a CMM machine, it was assumed that the part has been machined correctly. A Managing Director described it as following:
“Then you assume that the machine is correct, it is measured, it is straight, it is not crashed, and so on. Then it should be good, most likely.”
- R08, Managing Director
Two major reasons for not purchasing a CMM machine are likely because it is time-consuming to use and expensive to invest in. For the companies using CMM machines, the programming was indeed a very time-consuming task: from 2-3 hours up to multiple days, depending on the part’s complexity, degree of tolerances, and required documentation by the customer. More-over, there was no systematic collection of process feedback from the quality inspection back to CAD or CAM at any of the sample companies.
5.3
Value Stream Mapping
Among the three areas of improvements, it was decided to focus on the first one “Pain #1) Understanding and discussing the design intent, manufacturability, and tolerances are time-consuming, repetitive, and difficult to communicate efficiently”. This was decided due to it being the first step in the value chain and the other two pains being dependent on this one.
RESULTS & ANALYSIS
FIGURE 5.7: The sample companies’ workflows, excluding Company B, from receiving a drawing/model to confirming a work order and the effective time
spent on each activity, not including any waiting time.
First, the sample companies receive a drawing and, in most cases, a 3D model, often via email. After that, they review the drawing/model to make sure of its manufacturability. Next, changes in the model and/or the drawing are often communicated via mail or phone, except for Company A who also uses their helpdesk ticketing system. After that, the design changes may be discussed with the customer if necessary. For some companies, such as Company C & D, this happens with more or less every new order, while others, such as Company E & F, it happens more rarely. Company A estimates that for 10-20% of their received orders they need to have more extensive discussions with the customer, often face-to-face, but for all other or-ders, a request on the design changes via their ticketing system is normally enough. Company E makes design changes in the model directly and confirms the changes with the customer, while the others leave the modeling changes to the customer.
RESULTS & ANALYSIS
the order has been completed, if the part will be programmed using CAM, while the others request or receive the STEP file before this step. The last bar shows the total effective time spent on order handling, respectively for each company.
5.3.1 Analysis of Value Stream Mapping
Based on the value stream mapping presented in Figure 5.7, it can be noted that the most time-consuming steps in the initial workflow, from receiving a drawing/model to confirm-ing it, are “Discussconfirm-ing design changes with the customer”, “Analyzconfirm-ing drawconfirm-ing/model”, and “Developing order and send to customer.” The results imply that configuring is overall a very time-consuming task for the investigated sample companies. However, it is important to point out that the collected data in this value stream map are estimated averages, according to the re-spondents’ answers. This means that in some cases the effective time is less and in some cases, the effective time is much, much more. Thus, the margin of error must be considered relatively high.
It can also be noted in Figure 5.7 that the term “process planning” has likely been interpreted in different ways for the sample companies, which might have affected the respondents’ answers. Some of the sample companies viewed process planning as the creation of a physical process plan, while others would call the activity of CAM programming process planning. This is not unique to the sample companies of this thesis; Andersberg [31] described the terminology of process planning as “somewhat fuzzy” and several different definitions of the term currently exists [32]. For the sample companies, it is likely that some sort of process planning is happen-ing even if no physical process plan is created, however, the degree of structuredness of the process planning may vary between the companies. It is not unlikely that the process planning activity begins as soon as the sample companies start analyzing the 2D drawing and/or 3D model and are done throughout the workflow until the part will be manufactured, which can perhaps explain the respondents’ difficulty to estimate the time spent on each activity as many tasks blend together.
5.4
Value Delta
To illustrate the value of the time spent on the activities mentioned above, an estimation was calculated of what the total value delta could be worth. For this estimation, Company D’s data was used. They spent 325 minutes per work order (effective time) in total for the activities mentioned above. The following assumptions were made:
