The effectiveness of the proposed architecture is evaluated by the benefits this system provides the pilot company. Prior to the development and implementation of the pilot system in the pilot
Fig. 6.Interface 1 –production detail of operations in an order. Z.X. Guo et al. / Int. J. Production Economics 159 (2015) 16–28 23
manufacturing company, manual recordings were used to collect
production data. There is a computer operator to input daily job
tickets into the computer in each shopfloor of sewing department.
The production management failed to monitor the production
progress of each order in material supplying plants and clothing
production plants in a real-time and accurate manner. A large
amount of time is required to read and analyze daily summary
reports from each plant. The reports are not updated and unreliable because manual recordings and inputs often lead to data
errors. In addition, the production schedulers at the headquarters
assign each order to an appropriate plant based on experience and
subjective judgment, which usually generates inconsistent and
low-efficient solutions.
After implementing the pilot system for 6 months, performance was analyzed and evaluated by comparing the production
performance of the latest 3 months with that of corresponding
3 months in the previous year. Such comparison was conducted for
the following reasons:
(1) The system was confirmed as stable after several months of
operation in the pilot company. Therefore, production performance in the last 3 months represents the overall performance
of the system since implementation.
(2) Similar product types are produced in each plant in the same
months each year because of the seasonal characteristics of
clothing production. Comparison of the production performances in the same months for two consecutive years can
reduce production performance variations attributed to different product types.
(3) In the two selected periods, full-capacity production is
achieved, and the labor force is stable in each shopfloor.
Variation in performance can be attributed to the system after
elimination of factors that disturb production performance. Comparison of the two selected periods indicates that the pilot
company obtained the following tangible benefits:
(1) 25% Increase in production efficiency: production efficiency is
defined as the production workloads completed/number of
working hours, that is, the average production workloads
completed per hour. The production workload completed in
a period equals the total standard allowed minutes of all
operations completed within this period. The real-time and
accurate production data collected lead to a higher visibility
and transparency of production operations in distributed
manufacturing, which is a motivation for frontline operators
to reduce the possibility of loafing on the job. In addition, more
accurate progress monitoring of raw material production is
helpful to improve the effectiveness of production scheduling
decision-making and effective production scheduling decisions generated by the pilot system can improve production
efficiency as well.
(2) 12% Reduction in production waste: overproduction, defects,
and unnecessary inventory are identified as factors that lead to
production waste, which is defined as the total cost incurred
by these three factors. Overproduction costs are incurred
when the company produces more products than required
by a customer. Defect costs are incurred when bad production