Design of a Saw Cutting Machine for Wood and Aluminum

Jawad Ul Haq

, Ahmed Jawad Qureshi

and Mohamed Al-Hussein

Department of Mechanical Engineering, University of Alberta, Canada

Department of Civil and Environmental Engineering, University of Alberta, Canada

Keywords: Axiomatic Design, Table Saw, Programmable Logic Controller, Control System, Automation.

Abstract: The intensive use of wood in furniture, building, bridges, and of aluminum in transportation and construction,

underscores the economic importance of these building materials in North America. Power saws are very

useful tools for cutting and shaping such materials; however, they can cause serious hand injuries. In a table

saw operation for wood cutting, for instance, the operator’s hands are vulnerable as they are used to guide

pieces into the saw. In addition, the saw operator faces the risk of material being kicked back out of the saw

or of sustaining an eye or respiratory injury due to the presence of sawdust and other debris generating by the

operation of the saw. Meanwhile, aluminum cutting requires careful precaution and accuracy. The cutting can

be dangerous if not handled properly. The greatest challenge in this regard is securely holding the material

being cut. Furthermore, industrial requirements such as pneumatics and three-phase power supply preclude

the ready use of such machines on a domestic scale. The cutting capability of existing table saws is coupled

in such a way that it cannot cut both wood and aluminum. In this paper, a concept of a saw cutting machine

(SCM) is presented using Axiomatic Design to ensure design objectives such as safety, user comfort, usage

on a domestic scale and capability to cut both types of materials. In the presented case study, the mapping

from Customer Attributes (CAs) to Functional Requirements (FRs) and then respective Design Parameters

(DPs) resulted in an uncoupled design, in turn leading to a detailed mechanical design followed by the control

system, all based on the aforementioned design objectives.

1 INTRODUCTION

Forest products are a major contributor to the

Canadian economy. (Canada, 2013) In 2013,

production in the forestry sector contributed $19.8

billion to the economy. In a global context, Canada

has the world’s largest forest product trade balance.

The aluminum industry is another important sector of

Canada’s economy, with aluminum products export

valued at $10.8 billion in 2016, an increase of $211

over 2015; (Canada ranks third in aluminum

production in the world after China, and Russia).

The motivation for the development of the saw

cutting machine (SCM) described in this paper arose

out of a broader research initiative at the University

of Alberta (Canada) to develop a semi-automated

wood framing machine

Figure 1 and a semi-automated light-gauge steel

framing machine Figure 2. The primary objective of

both machine development projects is to support the

growth of panelized construction in North America’s

building construction sector. The structures of both

machines consist of aluminum extrusions that need to

be cut in different lengths and angles. The lab has to

outsource the cutting to third-party companies,

resulting in increased costs and delays of the machine

development program.

In order to overcome the aforementioned

challenges, the research team began investigating

SCM solutions with the design objectives of (1) the

ability to cut both wood and aluminum, (2) versatility

to be deployed in a lab or domestic scale without

three-phase industrial power supply or complex

pneumatics, (3) safety mechanisms to enable safe use,

Figure 1: Wood framing machine.

456

Haq, J., Qureshi, A. and Al-Hussein, M.

Design of a Saw Cutting Machine for Wood and Aluminum.

DOI: 10.5220/0006909704560464

In Proceedings of the 15th International Conference on Informatics in Control, Automation and Robotics (ICINCO 2018) - Volume 2, pages 456-464

ISBN: 978-989-758-321-6

Figure 2: Steel framing machine.

and (4) capable of angled cutting started and resulted

in a design discussed in the following sections.

In order to overcome the aforementioned

challenges, the research team began investigating

SCM solutions with the design objectives of (1) the

ability to cut both wood and aluminum, (2) versatility

to be deployed in a lab or domestic scale without

three-phase industrial power supply or complex

pneumatics, (3) safety mechanisms to enable safe use,

and (4) capable of angled cutting started and resulted

in a design discussed in the following sections.

The paper is divided into eight sections. Section 2

reviews the relevant literature with a focus on

potential safety hazards, Section 3 elaborates on the

design objectives and describes the research

methodology, Section 4 explains the implementation

of Axiomatic Design to form an uncoupled design,

Section 5 presents the mechanical design, and Section

6 the implementation of the control system to meet

the design objectives mentioned in Section 3. Section

7 describes the discrete-event simulation model of the

SCM, followed by Section 8, which summarizes the

research achievements.

2 LITERATURE REVIEW

Table saws are associated with more injuries than any

other type of woodworking tool.(Shields, Wilkins and

Smith, 2011) estimate that 565,670 table-saw related

injuries were treated in the United States during the

period 1990–2007. Injuries to fingers/thumbs were

the most common overall (86%—486,181 of

565,670). (Chung and Shauver, 2013) discuss

SawStop, a technology designed to stop the saw blade

when contact is made with skin, resulting in a small

cut rather than a serious laceration or amputation. A

few disadvantages associated with SawStop, though,

are that the force required to quickly stop the saw

blade damages the blade and brake beyond repair

such that they must be replaced each time the brake is

triggered; furthermore, the brake cartridges are blade-

specific; there are no brakes available for some

specialty blades; and brakes can only be used when

cutting nonconductive materials. (Graham and

Chang, 2015) provide a quantitative estimate of the

economic benefits of automatic protection systems

that could be deployed in new table saw products. The

general consensus among researchers is that the

benefits of automatic protection are likely to

outweigh the incremental costs of implementation

significantly. (Schwaneberg et al., 2012) discuss the

use of a LED-based sensor system to distinguish

human skin from work pieces. In this context, it is

thus of interest to investigate new technology to

automate the process by designing a machine using a

systematic method of design. (Farid and Suh, 2016)

Axiomatic design is a design method introduced by

Nam P. Suh. It consists of four domains: consumer,

functional, physical, and process. These domains are

interlinked in such a way that customer needs—

referred to as customer attributes (CAs)—in the

customer domain are transformed into functional

requirements (FRs) in the functional domain. FRs, in

turn, are satisfied by design parameters in the physical

domain. Product variables (PVs) are determined from

DPs in the same manner. Axiomatic design as

described above has been used in many fields, such

as software design (Suh and Do, 2000) and control

system design (Lee, Suh and Oh, 2001). (Negahban

and Smith, 2014) provide a comprehensive review of

discrete-event simulation in which the discrete-event

model of a system can be implemented using a

computer. This simulation-based approach can aid in

understanding the system under study in terms of

cycle time, utilization of different resources,

improvements in design, and production levels.

3 METHODOLOGY

The design objectives for the machine are as follows:

• Capable of cutting both wood and aluminum

• Can be used in a lab or domestic scale

• Ensures safety and operator comfort

• Can accommodate angled cutting

In general, the design of machines consists of

conceptual and detailed design processes. Overall the

factors which affect the most characteristics and the

cost are determined in the conceptual stage.

Axiomatic design is a design methodology to

systematically transform the CAs into FRs and then

respective DPs, and PVs. In this paper; Axiomatic

design is utilized in the conceptual design process that

further lead to detailed design. The FRs for the SCM

Design of a Saw Cutting Machine for Wood and Aluminum

457

are defined on the basis of CAs, and corresponding

DPs are selected. The detailed design is carried out on

the basis of decisions made in the conceptual stage.

Computer Aided Design (CAD) model of the SCM is

developed in SOLIDWORKS. Control system of the

SCM is realized on Programmable Logic Controller

(PLC) using Sequential Function Chart (SFC) which

is one of the IEC 61131-3 languages. In order to

estimate the performance of the machine, discrete-

event modelling technique is used. Arena input

analyzer by Rockwell automation is used to select the

distribution of each task in the simulation model.

4 AXIOMATIC DESIGN

Design process in Axiomatic design is top-down, in

which the initial concept is decomposed to design

details by zigzagging. The relationship between FRs

and DPs is given as

{FRs} = [A] {DPs} (1)

FRs are a minimum set of independent requirements

that completely characterize the functional needs of

the product in the functional domain. Each FR is

independent of every other FR at the time the FRs are

created. [A] is defined as the design matrix. When [A]

is the diagonal matrix, the design is called uncoupled

design which is ideal. When [A] is lower triangular

matrix, the design is called decoupled and preferred

in absence of uncoupled, while all the other designs

are called coupled. DPs are the physical variables in

the physical domain that characterize the design that

satisfies the specified FRs. When DPs do not take

their detailed physical form, the corresponding FRs

need further division. Based on DP

, FR

of the SCM

is decomposed into two sections as FR

3.1

and FR

3.2.

The FR

needs no further decomposition as DP

has

taken its detailed physical form.

The axiomatic design of SCM has three parts:

CAs, FRs, and DPs. At the beginning of the design

process, the needs of the customers (i.e., CAs) are

taken into account in order to generate the FRs and

then the DPs. The top-level design is given as

follows:

CA: Wood and aluminum cutting capability, safety,

user comfort, usage on a domestic scale, and angled

cutting capability.

: Carry saw, motors, sensors (electrical

components) inside a safe cabinet (mechanical)

: Programmable logic controller (PLC)-

controlled saw cutting machine

After the top level design, FRs and DPs are

decomposed and Table 1 illustrates the second level

FRs and DPs.

= Facilitate operator

= Industry power & pneumatics alternative

= Angle cut

= Safety

= Automation using stepper motors & Human

Machine Interface (HMI)

= Single phase power supply & force controlled

actuators

=Rotary table

= Sensors based mechanical assembly

The low level FRs and DPs decomposition is as

follows:

3.1

= Use automation

3.2

= Facilitate interaction with machine

4.1

= Use domestic power

4.2

= Use pneumatics alternative

6.1

= Incorporate safety measures against airborne

debris

6.2

= Make sure user is at a safe distance

3.1

= Motors

3.2

= Human Machine Interface

4.1

= Single phase power supply

4.2

= Forced controlled actuators

6.1

= Safety enclosure

6.2

= Ultrasonic sensors

Table 1: Initial design matrix.

FRs/DPs

1 Cut wood

2 Cut aluminum

1.1 Cutting RPM

1.2 Cutting

feed speed

1 Need to cut

wood

1.1 Use

cutting RPM

x x

1.2 Use feed

speed

x x

2 Need to cut

aluminum

2.1 Use

cutting RPM

x x

2.2 Use feed

speed

x x

ICINCO 2018 - 15th International Conference on Informatics in Control, Automation and Robotics

458

Table 2: Low level design matrix.

FRs/DPs

1.1 Apply cutting

wood RPM

1.2 Apply cutting

wood feed speed

2.1 Apply cutting

aluminum RPM

2.2 Apply cutting

aluminum feed speed

3.1 Stepper motors

3.2 Human Machine

Interface

4.1 Single phase

power supply

4.2 Force controlled

actuators

6.1 Safety enclosure

6.2 Ultrasonic sensors

1.1 Use cutting RPM

1.2 Use feed speed

2.1 Use cutting RPM

2.2 Use feed speed

3.1 Use automation

3.2 Ease interaction with machine

4.1 Use domestic power

4.2 Use pneumatic alternative

6.1 Incorporate safety measures

against airborne debris

6.2 Make sure user is at a safe

distance

The one obvious coupling which is not discussed for

this case study is the type of saw. A universal saw is

proposed to uncouple the design; although this will

compromise the quality of the cut in the case of

aluminum, it satisfies the design objectives and the

purpose for which the machine is being designed.

Feed speed =

× ×



(2)

Feed speed: inches per minute

RPM: revolutions per minute

Chip load: inches

The initial design matrix results in a coupled design

due to the fact that feed speed and RPM are related

(2). Feed speed and RPM have to be adjusted

according to the material being cut. The second

concern is that the number of FRs is greater than the

number of DPs. The first step towards uncoupling the

initial design is a permutation that results in a better

design but still a coupled one. To solve the issue of

feed speed and RPM coupling, a software solution is

used which is implemented on PLC that sets the

desired feed speed and RPM to cut the given material.

The second design issue of inequality in numbers of

FRs and DPs is addressed by adding more DPs to

make the design matrix square. The final uncoupled

design matrix is shown in Table 2.

5 MECHANICAL DESIGN

The CAD model of the SCM as shown in Figure 3 and

Figure 4 is developed in SOLIDWORKS, a solid

modelling computer-aided design tool that runs on a

computer. The machine design is flexible, it should

be noted, with regard to the length of material to be

cut. Depending on the length of the profile the table

can be attached along with a motor-controlled length

measurement unit, or the profile can be put directly

on the main cutting station. The force-controlled

actuators are used to clamp the piece firmly. A safety

enclosure protects against any debris or particle

hitting the operator while working, and the rotary

table is used to achieve the cut angle.

1. Table

2. Cutting length measurement unit

3. Main cutting station

4. Force-controlled actuators

5. Safety enclosure

6. Rotary table

Design of a Saw Cutting Machine for Wood and Aluminum

459

6 CONTROL SYSTEM

Machine control system is a collection of hardware

and software, designed to coordinate the output of

each individual component to achieve the desired

machine functionality.

Figure 3: SCM CAD model.

Figure 4: SCM main cutting station.

6.1 Process Flow

The process as shown in Figure 5 starts with the

manual loading of the wood/aluminum profile. A

human–machine interface (HMI) is used to obtain the

desired length and angle to be cut, followed by

clamping in which load sensors are used to apply the

required force to clamp the wood or aluminum being

cut. The saw motor waits for the safety enclosure to

come down and for the operator to move a safe

distance away.

6.2 Sequence of Operation

The sequence of operations consists of (1) loading;

(2) length and angle input; (3) clamping; (4)

engagement of safety enclosure and sensors for

operator’s safety; and (5) engagement of saw motor

and feed motor to cut material.

Loading is the manual operation in which the

operator picks a profile and places it in the designated

area of the machine. Once the loading is done, the

next step is to input material and cut specifications.

The HMI facilitates the interaction between the

operator and the machine. The information is inputted

to the machine by either of two methods. In the first

method, a computer numerical control (CNC) file

containing the complete information about the profile

is uploaded, and the machine reads the file in a

sequential manner to perform the operation. The CNC

file contains information such as the material,

coordinates, and angle to cut. In the second method,

the operator loads the profile and inputs the

information manually. Once the operator has inputted

the information, the machine executes safety

measures before carrying out the cutting operation. It

looks for a valid CNC file or coordinates to cut,

ensures by means of ultrasonic sensors that the

operator is at a safe distance, clamps the profile, and

engages the safety enclosure. If all the conditions are

met, then the PLC sends a command to the saw motor

to engage and perform the cut. Apart from these

safety checks, there are also emergency shutdown

(ESD) push-buttons which can be used to halt the

Figure 5: Process flow chart.

ICINCO 2018 - 15th International Conference on Informatics in Control, Automation and Robotics

460

machine in case of any abnormal scenario. To clamp

and

to replace the pneumatic system, feedback force-

controlled actuators are used. Based on the material

selected, the actuators apply the right amount of force

and the feed motor selects the desired feed speed to

cut the material. Once the material is cut, it is

unclamped in order for the operator to collect it.

6.3 Implementation of Control System

Automation of the sequence of operation is realized

by means of PLC as follows:

• Discrete inputs from proximity sensors for

wood/aluminum detection.

• Discrete inputs from limit switches for safety

and initial calibration.

• Analog inputs from load sensors to clamp

wood/aluminum.

• Analog inputs from ultrasonic sensors for

operator safety.

• Motor drive outputs to linear actuators for

clamping.

• Motor drive outputs to cut wood/aluminum at

desired angle and length.

• HMI to facilitate the automation process.

Once the hardware is known, next step is to select the

software to make harware operational. The PLC code

is developed in SoMachine, while the HMI code is

developed in Schneider Electric’s Vijeo Designer.

(Electric, 2018b) SoMachine is a software tool for

developing, configuring, and commissioning the

entire machine in a single software environment,

including logic, motion control, and related network

automation functions while Vijeo Designer is an HMI

configuration software. (Plaza, Medrano and Blesa,

2006) IEC 6113-3 standard is a global standard for

control programming that helps to improve software

quality. Ladder programming has several drawbacks,

including weak software structure and limited capacity

to handle complex data structures. (Jetley et al., 2013)

discuss the comparison of graphical IEC 61131-3

programs, asserting that it is easier to trace the error

with graphical languages as compared to textual.

6.3.1 Implementation of Code

The code for SCM is written in Sequential Function

Chart (SFC), which is one of the IEC 61131-3 langua-

ges. Each block in SFC has three portions: entry, main

body, and exit conditions. The program flows through

these portions in a sequential manner. The flow

between blocks is governed by transitions, which are

conditions the satisfaction of which drives the flow of

the program on to the next block as shown in Figure 6.

6.3.2 Implementation of HMI

Vijeo Designer provides great flexibility in designing

graphical user interfaces (GUIs), where the nature of

the operator’s interaction with the machine dictates

the design of the HMI. A well designed HMI aids the

operator in understanding the previous, ongoing, and

future tasks. It provides great advantages in terms of

providing a user-friendly interface even for users

without a relevant technical background, in which

warnings and emergency situations can be communi-

cated more efficiently by using bright colors to attract

the operator’s attention, and a single button can be

assigned multiple tasks providing more flexibility in

Figure 6: SCM code in SoMachine.

Figure 7: Material selection in Vijeo Designer.

Design of a Saw Cutting Machine for Wood and Aluminum

461

Figure 8: Operator input in Vijeo Designer.

terms of coding. The GUI implementation in Vijeo

Designer is shown in Figure 7 and Figure 8.

6.4 Ethernet/Ip Architecture

One of the complex tasks in the development of PLC-

based control systems is wiring. Having a relatively

small numbers of devices in a control system can

result in a complex wiring system which occupies

more space and is difficult to troubleshoot. (Electric,

2018a) Ethernet/IP uses two protocols for the

transport layer: Transmission Control Protocol (TCP)

and User Data Datagram Protocol (UDP). TCP is

acknowledged while UDP is unacknowledged

protocol. TCP is used for non-control messages while

UDP is used for Input/output (I/O) messages. Tested

validated document and architecture (TVDA)-based

Ethernet/IP improves efficiency in the design and

planning phase. Based on inputs/outputs described in

Section 6.3 Ethernet/IP architecture is used for the

machine described in this paper due to the following

advantages:

• Ability to access the machine from anywhere,

anytime via Ethernet for remote monitoring.

• Flexible in terms of adding an Ethernet/IP

adaptor at any time.

• Efficient in terms of device integration and

configuration, and the architecture can easily

be modified.

With embedded Ethernet/IP communication, a

PLC can communicate with 16 slaves in 10 ms. The

Ethernet/IP architecture for the SCM is given in

Figure 9.

7 DISCRETE EVENT

SIMULATION

Discrete-event simulation describes a process with a

set of unique, specific events in time. Arena by

Rockwell automation is used in the present research

to build the SCM model with its key performance

parameters such as cycle time and operator

utilization. The model as shown in Figure 10 reads a

CNC file that contains information about a profile,

such as material, cut coordinates, and cut angle, in a

sequential manner. The task times and triggers are

assumed to provide the basis for statistical analysis

and hypothesizing distribution.

Figure 9: SCM Ethernet/IP architecture.

7.1 Model Discussion

A discussion of the simulation model is given in this

section. In Arena, a model is built using a “process”

module that holds the “entities” for a specific period

of time. The entities flow through different process

modules to provide a valuable insight into machine’s

key performance indicators at the end of the

simulation. The model for the SCM reads a

spreadsheet extracted from a CNC file and scans the

total number of cutting operations in the file prior to

processing. It then generates entities equal to the

number of cutting operations. The “Hold Entity”

module holds the next entity, which is the next cutting

operation, until the previous entity, which is the

previous cutting operation being processed by the

model, finishes. The “Decide Operation” module

decides the material on the basis of a string variable

that looks for either “WD” for wood or “AL” for

aluminum in the file and then guides the respective

entity to pass through the modules designated for the

respective material. The “Load Profile” and “Collect

ICINCO 2018 - 15th International Conference on Informatics in Control, Automation and Robotics

462

Piece” modules share a common resource, which is

the operator. The “Length” and “Angle” module task

times, it should be noted, are dependent on the

coordinates and proportional to the cut length and

angle in the CNC file. The model consists of

following main modules sections (1) initial

calibration; (2) CNC file reading; (3) aluminum

cutting; (4) wood cutting; and (5) ending. (1) accounts

for the time taken in homing the motors and initial

system delays when the machine is powered on, (2)

deals with reading of the CNC file and deciding the

operations accordingly, (3) accounts for the time

taken in cutting the aluminum, (4) accounts for the

time taken in cutting wood, and (5) indicates when all

the operations on the profile are done.

7.1.1 Case Study

To illustrate the effect of different profiles with

different cut and angle coordinates on the key

performance indicators, for instance, cycle time and

utilization of the operator, Table 3 shows the

summary of results obtained from the model. For the

profile case studies as illustrated in Table 3, the

simulation model generates a total of five entities, as

there are five cutting operations at time zero. The

“Hold Entity” module holds the next cutting

operation until the previous entity or cutting operation

exits the model, and sends a trigger to the hold

module through the signal module. The simulation

keeps running until all the entities generated have

exited the model.

Figure 10: SCM simulation model.

Table 3: Simulation results summary.

Profile Dimensions

WxHxL

(inch)

Material Cut Lengths

(inch)

Cut Angle

(θ)

Average

Cycle Time

(minute)

Average

Operator’s

Utilization

(%)

Profile 1

1.57x1.57x78.74 Aluminum 12,24,48 45,60,0 3.3 81

Profile 2

1.5x3.5x78.74 Wood 12,36 0,0 2.1 70

Design of a Saw Cutting Machine for Wood and Aluminum

463

8 CONCLUSION

The traditional method for cutting wood using a table

saw involves a stationary saw motor in which the

wood is fed through the saw by hand. This approach

entails serious safety hazards. On the other hand,

aluminum cutting requires extra precaution and

careful craftsmanship to ensure an accurate cut, and

the cutting can be dangerous if not executed properly.

Given the inherent risks of conventional sawing

practice, limitations of cutting both materials,

benefits of automation and to support panelized

construction, in this paper Axiomatic design theory is

applied for investigating the problems of the present

table saws and for designing an uncoupled new one.

As a result of maping from functional domain to

physical domain, the feed speed and RPM for wood

and aluminum cutting found to be coupled. A

complete control system strategy from defining the

process flow to its full implementation was crafted to

meet the design objectives and based on the analysis

an uncoupled design of saw cutting machine is

introduced. Discrete event modelling is employed to

estimate the performance of the machine and

implication of different sizes of profiles. The

simulation results provide valuable insight into

machine’s key performance indicators, for instance,

cycle time and operator’s utilization.

REFERENCES

Canada, N. R. (2013) Forest and Aluminum Industry.

Available at: http://www.nrcan.gc.ca

Chung, K. C. and Shauver, M. J. (2013) ‘Table Saw

Injuries’, Plastic and Reconstructive Surgery, 132(5),

p. 777e–783e. doi: 10.1097/PRS.0b013e3182a3bfb1.

Electric, S. (2018a) Industrial Ethernet. Available at:

https://www.schneider-electric.com/en/product-range-

presentation/1106-industrial-ethernet?parent-category-

id=2400&parent-subcategory-

id=2410&filter=business-1-Industrial Automation and

Control.

Electric, S. (2018b) SoMachine. Available at: https://www.

schneider-electric.ca/en/product-range-presentation/22

26-somachine#tabs-top.

Farid, A. M. and Suh, N. P. (2016) Axiomatic Design in

Large Systems. doi: 10.1007/978-3-319-32388-6.

Graham, J. D. and Chang, J. (2015) ‘Reducing the risk of

injury from table saw use: The potential benefits and

costs of automatic protection’, Risk Analysis, 35(2), pp.

307–317. doi: 10.1111/risa.12258.

Jetley, R., Rath, A., Aparajithan, V., Kumar, D., Prasad, V.

and Ramaswamy, S. (2013) ‘An approach for

comparison of IEC 61131-3 graphical programs’, IEEE

International Conference on Emerging Technologies

and Factory Automation, ETFA. doi: 10.1109/

ETFA.2013.6647938.

Lee, K. D., Suh, N. P. and Oh, J. H. (2001) ‘Axiomatic

design of machine control system’, CIRP Annals -

Manufacturing Technology, 50(1), pp. 109–114. doi:

10.1016/S0007-8506(07)62083-6.

Negahban, A. and Smith, J. S. (2014) ‘Simulation for

manufacturing system design and operation: Literature

review and analysis’, Journal of Manufacturing

Systems. The Society of Manufacturing Engineers,

33(2), pp. 241–261. doi: 10.1016/j.jmsy.2013.12.007.

Plaza, I., Medrano, C. and Blesa, A. (2006) ‘Analysis and

implementation of the IEC 61131-3 software model

under POSIX Real-Time operating systems’,

Microprocessors and Microsystems, 30(8), pp. 497–

508. doi: 10.1016/j.micpro.2006.06.001.

Schwaneberg, O., Steiner, H., Bol\’ivar, P. H. and Jung, N.

(2012) ‘Design of an LED-based sensor system to

distinguish human skin from workpieces in safety

applications’, Appl. Opt., 51(12), pp. 1865–1871. doi:

10.1364/AO.51.001865.

Shields, B. J., Wilkins, J. R. and Smith, G. A. (2011)

‘Nonoccupational Table Saw-Related Injuries Treated

in US Emergency Departments, 1990–2007’, The

Journal of Trauma: Injury, Infection, and Critical

Care, 71(6), pp. 1902–1907. doi: 10.1097/TA.0b

013e3181b28ad3.

Suh, N. P. and Do, S.-H. (2000) ‘Axiomatic Design of

Software Systems’, CIRP Annals - Manufacturing

Technology, 49, pp. 95–100. doi: 10.1016/S0007-

8506(07)62904-7.

ICINCO 2018 - 15th International Conference on Informatics in Control, Automation and Robotics

464