You are on page 1of 4

Available online at www.sciencedirect.


Procedia CIRP 59 (2017) 156 – 159

The 5th International Conference on Through-life Engineering Services (TESConf 2016)

Study on quality improvement of machine tools

Makoto Fujishima1, Masahiko Mori1, Kimiyuki Nishimura1, Katsuhiko Ohno1
1 DMGMORI Co,. Ltd., Nagoya, Japan,

* Corresponding author. Tel.: +81-52-587-1811; fax: +81-52-587-1818. E-mail


Quality management of machine tools is the key to reducing downtime of customers’ machines and the manufacturers’ service cost. This paper
introduces DMG MORI’s systematic workflow from designing, procurement, manufacturing, and acceptance inspections to the delivery of
machines, which contributes to improvement of machine quality. Two quality management systems, one for before shipment and the other for
after shipment, are described in this paper. Various workflow systems, data analysis, remote monitoring, continuous Kaizen approaches are
integrated in the systems. However, defects of machines cannot be completely eliminated, so proactive maintenance and strong service
networks are vital for business success. Especially, proactive maintenance is favorably accepted in the industry as a solution to minimize
machine downtime and the repairing cost. This paper introduces various sensing technologies to improve the functions of proactive

2016 The
The Authors.
Authors. Published
Published by Elsevier
by Elsevier B.V.
B.V. This is an open access article under the CC BY-NC-ND license
Peer-review under responsibility of the Programme Committee of the 5th International Conference on Through-life Engineering Services
Peer-review responsibility of the scientific committee of the The 5th International Conference on Through-life Engineering Services (TESConf 2016)

Keywords: machine tools; smart manufacturing; sensing

1. Introduction DMG MORI CO (DMG MORI in Japan) and 130,000

machines of DMG MORI AG (DMG MORI in Germany) are
Until recently, machine tools utilizing the state-of-the-art in operation. Moreover, according to the survey, 230,000
technologies such as 5-axis machines and mill turn centers and companies are potential customers of DMG MORI.
the machining methods using those machines were mainly Large-scale manufacturers have installed hundreds of
studied in research and development of the machine tool thousands of machine tools at customers’ factories worldwide,
industry. However, new approaches to technical development so they need to provide meticulous services to customers to
have been drawing attention lately, e.g. various technologies keep those machines constantly operating. If a machine tool of
such as machine tools, peripheral equipment, control units, a customer breaks down, service engineers must visit the
and applications are combined. To be more specific, systems customer’s factory to repair the machine, and the production at
utilizing the IoT technology and the concept of Industry 4.0 the customer’s factory stops until the machine recovers. In
and the latest machining methods such as additive most of such cases, customers demand that the manufacturer
manufacturing are incorporated into new solutions. The should recover the machine as promptly as possible.
authors have been working to employ the above-mentioned Furthermore, the average life span of machine tools is
technologies to machine tools. However, among customers’ considered to be as long as 15 years and maintenance services
demands, having machines that do not break down once they are required during the same period. To provide maintenance
are installed and start operating at their factories are still the services to customers for such a long period, it is necessary to
most important. carry spare parts with short life cycles such as electrical parts
The number of DMG MORI’s machine tools installed at in stock or modify machines so that the latest spare parts can
customers’ factories is approximately 200,000 machines of be used.

2212-8271 © 2016 The Authors. Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license
Peer-review under responsibility of the scientific committee of the The 5th International Conference on Through-life Engineering Services (TESConf 2016)
Makoto Fujishima et al. / Procedia CIRP 59 (2017) 156 – 159 157

The key to providing customers with machine tools which Fig. 1 shows how the processes of the manufacturing
do not break down is to improve quality during the production management, quality management, and acceptance inspection
processes. This is the common target for most of the divisions are combined. These processes and the quality of machines
in the entire plant including development and manufacturing tools are all managed at one central system by computers. Fig.
divisions. The following are the items to be achieved. 2 shows the flow of assembly work in a vertical space
1. Common items utilizing the 3D model and BOM system to create standard
Lean manufacturing procedures for assembly work and test procedures.
PDCA cycle 2.1 Quality management before shipment
2. Design section Quality inspections are conducted to detect problems
Design review including potential problems at the factories before shipment
Design verification to prevent problems from occurring after machine tools are
Customer point of view evaluation delivered to the customers’ factories. Fig. 3 shows the flow of
3. Manufacturing section the product inspections including the positioning accuracy test,
Before shipment quality management cutting test, product inspection, running test, and shipment
Kaizen inspection.
Inspection In the quality inspection processes actions for improvement
4. Procurement section are continuously taken, contributing to the quality
Supplier management improvement of machine tools.
Acceptance inspection 㻵㼚㼢㼑㼚㼠㼛㼞㼥㻌 Digital picking 㻼㼞㼛㼏㼑㼟㼟㻌
Process management 㼙㼍㼚㼍㼓㼑㼙㼑㼚㼠
Cost management 㼙㼍㼚㼍㼓㼑㼙㼑㼚㼠
VPS䠖Virtual product
Parts preparation
Delivery management simulator Operation

If these items are rigorously pursued, drastic improvement in

the quality can be expected, defects on machine tools cannot Acceptance
be completely eliminated. So efficient and highly-qualified
Web order
services are important, and the following two solutions are the
key to providing such services. Test procedure
NET Inspection monitor

(1) Detect failures before parts broken 㻻㼞㼐㼑㼞㼞㻌 㻽㼡㼍㼘㼕㼠㼥㻌

㼙㼍㼚㼍㼓㼑㼙㼑㼚㼠 㼙㼍㼚㼍㼓㼑㼙㼑㼚㼠
Preventive maintenance with sensing technologies Fig.1 Smart manufacturing at Iga facility of DMGMORI
(2) Repair as soon as possible after failure
Development Preparation Manufacturing Quality
Remote monitoring and maintenance system management
Strong service network 3D CAD DATA 䠄Virtual Product Simulator䠅
Check sheet
Parts supply system Traceability

This paper introduces approaches to the above-mentioned Interference check reader
contents as a machine tool manufacturer.
Standard Transfer
Process managem
me t
2. Quality management at the factory
DMG MORI has 13 main factories in seven countries. This Operation
global production system, with factories in Europe, the U.S.A.,
and China, as well as four factories in Japan and three in ORiN
Feed back
Germany, reduces risks of volatile currency exchange rates,
and enables optimal machine tool delivery for the customers Fig. 2 Utilization of Design data
worldwide in terms of transportation costs and lead time.
In-process inspection Positioning accuracy test Cutting test
Additionally, high accuracy magnescales and laser scales are
ISO standard test Cutting accuracy test
Self inspection by
produced in Isehara, and Taiyo Koki's grinding machines are the assembly
produced in Nagaoka. At the factory in Iga, the largest one
among the 13 main factories, medium-size and large-size
machine tools are manufactured. The quality management
divisions in Iga, which consists of two divisions: the before Shipment Shipment inspection Product inspection
100-hours running operation
shipment quality management division and after shipment Inspection of packaging Overall running operation of 100 specifications,
and bundled items at hours in total, covering all functions,
quality management division. The before shipment quality shipm
shipment roces
o ce
c e
se embedded software
and safety
management division is responsible for inspecting machines
before shipment to provide customers with high-quality
machine tools, whereas the after shipment quality
management division is responsible for improving the quality
of machine tools after they are installed at customers’ factories. Fig.3 Product inspection
158 Makoto Fujishima et al. / Procedia CIRP 59 (2017) 156 – 159

3. Sensing technologies for machine tool

The paper has explained quality management and services
to be provided after shipment so far. As a next challenge, the
paper introduces preventive maintenance and predictive
New Tolerance(2014 maintenance [2], [3] in which sensing technologies are utilized.
䡚䠅 Many sensors have been effectively used lately in various
fields such as consumer appliances, health care, transportation,
and businesses. Numerous sensors actually function these days,
many of which were once thought to be inconceivable to be
put into practice. Consequently, the production volume of
sensors increased, while production costs per unit decreased.
According to the 2013 TSensors Summit, the volume in the
mobile sensor market grew exponentially, exceeding annual
Fig. 4 Accuracy improvement process increases of 200% between 2007 and 2012. Visionary
organizations foresee the number of sensors will reach trillions
Fig. 4 shows the sampling of measurement results of the X- within the next decade [4], exceeding the sales of billions in
axis positioning accuracy of horizontal machining centers. The 2012 due to the recent growth of the demands for sensors..
accuracy of 80% within the allowable range is constantly More and more sensors will be utilized for machine tools in
achieved thanks to daily efforts to take actions for accuracy the near future. Members of CIRP have been studying various
improvement. The allowable range of accuracy is reviewed types of monitoring technologies by using sensors [5] , [6],
every year and an even more challenging target is set and determined that some of the technologies will be effective,
depending on the measurement result. if they are put into practice. The goal of this study is, therefore,
to develop a test machine equipped with various sensors and
2.2 Quality management after shipment utilizing practical applications in the industry by employing
In the quality management after shipment, problems which sensing technologies that were studied in the past.
have occurred at customers’ factories are managed, so solving In this study, acceleration sensors embedded in the spindle
those problems and taking countermeasures to prevent unit were used to measure chatter vibration, collision impact,
recurrence of the same problems are important. If a problem and abnormal bearing vibration. A coolant level sensor,
occurs at a customer’s factory, the service engineer in the developed by ourselves, is advantageous in terms of cost and
region issues the product problem report system (PPR). The performance. Electric current and voltage sensors were used to
After Shipment Quality Management Group receives the measure energy consumption of the entire machine tool.
report, and forwards it to the related division(s) with Temperature sensors embedded in the spindle unit and table
instructions about what need to be done. When a critical were used to compensate thermal displacement or diagnose
problem occurs or the same type of the problem occurs on spindle bearings. Motor power and spindle speed obtained
multiple machines, the After Shipment Quality Management from the NC unit were utilized as well.
Group instructs to change design or change the procedure in Machine tool input-output (I/O) signal lines are usually
the manufacturing process to prevent recurrence of the same connected to the Programmable Logic Controller (PLC). I/O
problem. signals are mostly digital binary at DC24V. Analogue signals
Remote monitoring is efficiently utilized in the quality can be received by the PLC, but a generic PLC is not suitable
management after shipment. Approximately 10,000 machine to receive high-speed analogue signals in sensing technology
tools are connected to the DMG MORI’s network and these studies. The sampling rate required for acceleration sensors
machine tools are monitored remotely [1]. When a problem exceeds several kHz, which is several fold of chatter vibration
occurs at DMG MORI’s customer’s factory, detailed frequency. Since generic PLC control cycles commonly
information about the machine such as the diagnosis result by perform only in the millisecond range, a platform that allows
CNC and the alarm history can be checked at offices of DMG machine tool manufacturers to utilize sensor data without
MORI to promptly solve the problem (Fig. 5). limitations of the PLC or NC unit was developed [7].
Customer’s Factory Machine Tool The backbone route of the sensing network was structured
Problem mobile phone Manufacturer’s Service
with the 100BASE-TX Ethernet. Because sensors cannot be
connected to the Ethernet directly, four types of interface
boards for sensor signal inputs and Ethernet outputs were
Diagnostic requests can be automatically sent by
machine or manually sent by machine operator.
Problem is diagnosed and repaired developed.
Screen information
The Data Acquisition FFT Board (DAQF) has three
(Data is compressed) acceleration sensor interfaces and two temperature sensor
interfaces. In addition, the real time FFT function is embedded
Operate and monitor CNC screen
One screen information can be
in this circuit board. The Data Acquisition Temperature Board
displayed in 3sec
(DAQT) is a general A/D converter board with the voltage
input terminal and thermistor interface. The Electrical Power
Monitor Board (EPM) is connected to electrical current
Fig. 5 Remote monitoring system sensors and voltage sensors, which were placed at three-phase
Makoto Fujishima et al. / Procedia CIRP 59 (2017) 156 – 159 159

The coolant level in the tank can be easily detected with this
ATC, Magazine Chattering control Balance monitoring/ control
sensor. That is, monitoring the clean tank and dirty tank
Preventive Maintenance 䞉XYZ Vibration sensors
䞉 Motor load and torque 䞉Motor load 䞉SP Balancing data enables us to predict when the filter between the tanks needs
䞉Bearing temp. and vibration data
B Axis automatic
rotation adjustment
cleaning or replacing. The timing is when the difference
䞉 Servo data from motor Hydraulic management
䞉 Pressure between coolant levels of both tanks becomes higher than a
䞉 Controller alarm
Power monitor 䞉Contamination of oil
certain level.
䞉Power consumption
(Voltage & current) Oil cooler management Reducing the power consumption of a machine tool is one of
Remote monitoring 䞉 M/C body temp.
䞉Alarm signal 䞉 Oil Temp. the most important themes. The authors have studied power
䞉 Atmosphere temp.
䞉Operation monitor
䞉sensing data
䞉 Motor load saving and experienced that power consumption was actually
䞉 Controller alarm

Automatic servo Coolant quality

reduced on actual machine tools by employing some of the
tuning Distortion
䞉Servo data from motor
䞉Vibration data
䞉Coolant level
technologies that were studied. The next step is to monitor and
䞉Vibration data 䞉Load cell data 䞉Temp. data 䞉 Water level and temp. data
䞉 Contamination level visualize power consumption of each machine tool. A digital-
computing-type power meter was developed for this. The
Fig.6 Sensing items for machining center voltage sensor and electric current sensor were attached to the
power supply. Aside from calculating electrical power three-phase power input area so that samples of the voltage
consumption, this function monitors power supply quality. and electric current are collected simultaneously at a high
The EPM is also connected to the coolant level sensor. Fig. 6 speed, analog data is converted to digital data, and
shows all the sensing items which could be mounted on a instantaneous power is calculated by multiplying the data with
machining center, e.g. sensors for the spindle, table, ATC, the microprocessor(Fig.8). The result is processed by the
magazine, coolant, power consumption and so on. The coolant averaging procedure for a period of one cycle (T), the active
level sensor and power consumption sensor are introduced. power P is calculated, and the active power P is displayed on a
A new approach for the measurement of the coolant level is HMI via the Ethernet. This system measures power
introduced here. This system uses a capacitance-type original consumption and detects abnormal power consumption as well,
level sensor shown in Fig. 7. The electrodes of this sensor fall which enables us to determine machining processes and
into nine areas. Each electrode pair has different capacitance predict potential problems for machine tools.
values depending on whether the electrode is in the air or Though many sensor technologies have been widely used on
submerged in coolant. machine tool, the security of such application is still a concern.
The collected capacitance values were converted to voltage This includes the integrity of sensing data transmission,
values and transferred to the EPM through the RS485 customer specific data protection and reliable sensing
interface. Nine electrode pairs as the sensor device and the acquisition system to prevent harmful attack. The security
control circuit are on one circuit board, which makes the issue needs further investigation, and this paper does not cover
sensor cost less expensive than a currently available level the topic.
4. Conclusion
(1) Smart manufacturing enables us to efficiently manage the
quality of machine tools before shipment in the factory.
(2) Remote monitoring enables us to efficiently  perform
maintenance work for customers’ machines after shipment.
(3) Sensing enables us to improve functions of preventive


[1] Mori, M. Fujishima, M. Komatsu, Bingyan Zhao, Yadong Liu, 2008,

Fig. 7 Level sensor and inner circuit Development of Remote Monitoring and Maintenance System for
Machine Tools, Annals of the CIRP, 57/1
Machine PCB board
[2] Jay Lee, Machine performance monitoring and proactive maintenance in
Sensor data Ethernet computer-integrated manufacturing: review and perspective, International
Journal of Computer Integrated Manufacturing, 4/3/2007
0䠉440V 0䠉6V [3] Cunha, P.F. Caldeira Duarte, J.A. Alting, L. 2004, Development of a
ADC 㽢 P1
productive service module based on a life cycle perspective of
maintenance issues, Annals of the CIRP, 53/1, 13-16
Data collection
[4] N.N, 2013, The TSensors Roadmap,
Ptotal = P1 + P2,
i1 Factory server
5 [5] H.C. Möhring, K.M. Litwinski, O. Gümmer, 2010, Process Monitoring
6 i2
with Sensory Machine Tool Components, CIRP Annals, 59/1:383-386
V1 V2
[6] R. Teti, K. Jemielniak, G. O'Donnell, D. Dornfeld, 2010, Advanced
monitoring of machining operations, CIRP Annals, 59/2:717-739.
[7] Fujishima, M., Ohno, K., Nishikawa, S., Nishimura, K., Sakamoto, M., &
Fig.8 Power consumption monitoring system Kawai, K. 2016. Study of sensing technologies for machine tools. CIRP
Journal of Manufacturing Science and Technology.