This article developed a Manufacturing Resource Planning (MRP)II for a company specializing in the medical garment industry. The resources of the company had not been planned and controlled effectively, leading to high production costs and low on-time delivery, affecting the competitive edge of the company. The research effectively planned all the company’s resources to reduce production costs, increase the on-time delivery rate, and finally improve the competitive edge of the company. The paper integrated all MRPII functional modules of the companies, including Demand Management (DM), Production Planning (PP), Master Production Schedule (MPS), and Material Requirement Planning (MRP). The paper applied the available to promise (ATP) model to help process customer orders effectively and applied operations research models in PP to minimize production costs.
manufacturing resource planning (MRP)IIdemand managementforecast consumptionproduction planningmaster production schedule (MPS)available to promise (ATP)material requirement planning (MRP)Introduction
Since the beginning of 2020, the Covid-19 epidemic has had a major impact on most industries. However, demand was increasing for the medical garment industry, opening up great opportunities for businesses in this industry. Instead of being able to take advantage of the opportunity to increase customers, MZ company (a company in medical garment industry in Vietnam) (MZC), lost a significant number of orders.
The 5-Why method was applied to find the initial causes of the lost-order problem. The initial causes were high production costs and low service levels. The company faced cost problems. An increased cost in resources leads to an increased cost in production that affects their market share. Manufacturing resources needed to be used effectively to cut costs.
The company’s current service level, ability to meet orders on time, was quite low. The average order delay rate in the first 10 months of 2020 is about 64.8%. A fishbone diagram combined with the brainstorming method was used to analyze the main causes of late orders, as shown in Figure 1.
Fishbone diagram for the possible causes of late orders.
Using the failure mode and effects analysis (FMEA) method to identify and prioritize the causes of late orders in the production of the company. The result with risk priority number (RPN) percentage P(%) and RPN cumulative percentage CP(%) was shown in Table 1.
Analyzing and prioritizing possible causes.
No.
Causes
Codes
Severity (SEV)
occurrence (OCC)
detection (DET)
RPN
P
CP
1
Inappropriate production plan
A
8
8
7
448
22.63
22.63
2
Improper production schedule
B
8
7
6
336
16.97
39.60
3
Unspecific forecasting models
C
7
7
6
294
14.85
54.44
4
Incorrect material plans
D
8
6
5
240
12.12
66.57
5
Wrong order promising
E
8
5
4
160
8.08
74.65
6
Inconsistent Bill of Materials (BOM)
F
6
5
4
120
6.06
80.71
7
Inefficient capacity plans
G
8
3
4
96
4.85
85.56
8
Inappropriate purchasing plans
H
6
4
3
72
3.64
89.19
9
Inexperienced workers
J
5
2
6
60
3.03
92.22
10
Long waiting time
J
5
4
3
60
3.03
95.25
11
Lack of capital
K
7
3
2
42
2.12
97.37
12
Machines unbalance
L
3
4
2
24
1.21
98.59
13
Unsuitable HR schedule
M
4
2
2
16
0.81
99.39
14
Long changeover time
N
3
2
2
12
0.61
100.00
The Pareto chart for the main causes of late orders was shown in Figure 2.
The Pareto chart for the causes of late orders.
The six main causes of late delivery with an RPN cumulative percentage of more than 80% were inappropriate production plans, improper production schedules, unspecific forecasting models, incorrect material requirement planning (MRP), wrong order promising, and inconsistent bills of materials. The six causes with corresponding solutions were shown in Table 2.
Causes analysis and solutions.
No.
Causes
Solutions
Functional modules
1
Inappropriate production plan
Establish an optimal production plan to meet customer needs
Production Planning (PP)
2
Improper production schedule
Establish, adjust, and control the production schedule efficiently to meet the production plan.
Master Production Schedule (MPS)
3
Unspecific forecasting models
Establish forecasting models that matched the historical data
Demand Management (DM)
4
Incorrect material plans
Establish a material requirement plan to meet the MPS
Material Requirement Planning (MRP)
5
Wrong order promising
Develop Available to promise model in MPS
MPS
6
Inconsistent BOM
Control BOM in MRP
MRP
The solutions led to a manufacturing resource planning (MRP)II, which consists of four functional modules as shown in Table 2. The MRPII system was designed to reduce late orders and then increase the customer service level. The system also reduced resource costs, thereby helping to solve the problem of lost orders for the company. The MRPII system was constructed for the SS product family of the company, including six models: SS1, SS2, SS3, SS4, SS5, and SS6.
Literature review
According to (1), the MRPII is a system for effective planning of all resources of a manufacturing company. It is made up of a variety of interlinked functions such as Demand Management (DM), Production Planning (PP), Master Production Schedule (MPS), MRP, Capacity Requirement Planning (CRP), and Vendor Requirement Planning (VRP). The MRPII is an approach to managerial planning, execution, and control of productive activity. It integrates, in a feedback manner, the forecasting of demand, PP, scheduling and control, and purchase planning and control.
Amad-Uddin et al. designed a bespoke MRPII system for an small and medium enterprise (SME) company with the critical modules DM, MPS, MRP, and CRP (2). Phong et al. developed an MRPII system in the electric generator manufacturing industry with modules of DM, PP, MPS, and MRP (3).
Purna Chandra Padhan used Seasonal Autoregressive Integrated Moving Average (SARIMA), a time series forecasting model, in forecasting cement productions. The forecasting performance of various competing models was evaluated through forecast accuracy criteria mean absolute percentage error (MAPE) (4). Emiro Antonio Campo et al. proposed and implemented an aggregated PP model to provide optimal strategies in the medium term for a textile company. Linear programming was used to minimize total costs associated with labor and inventory levels. The model took into account characteristics associated with fabric contraction, wastes in the process, the efficiency of new employees, and training requirements (5).
Jia-Nian Zheng and Chen-Fu Chien developed a system for excelling enterprise resources for light-emitting diode (LED) manufacturing that optimizes chip procurement and PP by linear programming (6). P.J. Weeda constructed a stochastic model for the reduction of the initial forecast in the master schedule of an MRP system during the progress of time by the acceptance of customer orders (7).
Raqeyah Jawad Najy found out that an overstated MPS caused raw materials and work in process (WIP) inventories to increase and led to missed due dates. He used rough cut capacity planning (RCCP) to validate the MPS with respect to available capacity (AC) (8). Juin-Han Chen and Chin-Tai Chen proposed a two-phase order promising process in which available to promise (ATP) was first reserved in phase I to meet the most important demand. Customer orders were promised in Phase II according to the time-phase manufacturing resource supply calendar and restricted via the ATP in Phase I (9).
Teeradej Wuttipornpun et al. developed a practical finite capacity MRP system based on the needs of an automotive-part manufacturing company in Thailand. The proposed system offered a good trade-off between conflicting performance measures and resulted in the best weighted average performance measure when compared with conventional forward and forward-backward finite capacity scheduling systems (10).
Research methodology
The research methodology for developing the MRPII system includes the following steps.
Step 1: Develop DM module
Step 2: Develop PP module
Step 3: Develop MPS module
Step 4: Develop MRP module
Develop demand management (DM) module
This module has to achieve the objective of accurately forecasting customer demand. The DM module has two inputs: the long-term demand, or historical demand, and short-term demand, or customer order. Historical demand data in the product family will be used to make a demand plan for the PP module by suitable forecasting models. Forecasting models are also used to forecast product mix percentages for each product in the product family. Customer orders will go through the DM module to the MPS.
Forecasting is done according to the following steps:
Determine forecast parameters
Collect data
Eliminate outliers in the data
Select appropriate forecasting models
Forecast according to the best model.
Forecast parameters include forecast objects and forecast times. The forecast object can be a product family or individual product. The forecast time is 1 year with a cycle of 1 month. Past data is collected to forecast the forecast object. The data is processed, and outliers are eliminated to ensure the reliability of forecast data. Data patterns are analyzed to select appropriate forecast models. The best forecast model is selected according to forecast errors.
Develop production plan (PP) module
The PP module receives the demand plan from the DM module and constructs the production plan of the product family to minimize the cost of using the resources and constraints on resource capacity (1). Production plans are constructed by using operations research models in (11).
The PP steps are as follows.
Identify resource alternatives.
Construct the model
Collect data, estimate model parameters
Determine the production plan.
Resource alternatives used to meet demand in the case include regular-hours production, overtime production, inventory, and back-order. The model is constructed by defining model indexes and parameters, thereby determining the objective and constraints.
The index and parameters used in this case are defined as shown in Table 3.
The model index and parameters.
The index and parameters
Meaning
t
Time period (t = 1÷12)
D(t)
Demand in time period t
r
Regular hour production cost per unit
RC(t)
Regular hour production capacity
o
Overtime production cost per unit
OC(t)
Overtime production capacity
h
Inventory cost per unit per period
MI
Maximum inventory
II, EI
Initial and ending inventory
b
Backorder cost per unit per period
MB
Maximum backorder percentage
IB, EB
Initial and ending backorder
The decision variables used in this case are defined in Table 4.
The model variables.
Variables
Meaning
X(t)
Number of products produced in a regular hour in period t
Y(t)
Number of products produced in overtime in period t
I(t)
Number of products in inventory in time period t
B(t)
Number of products in back order in time period t
To minimize the total resource cost, the objective function is defined as follows.
Min C
C=∑t=1T(rX(t)+oY(t)+hI(t)+bB(t))
The constraints are constructed as follows with t = 1÷12.
After the model is built, data is collected to estimate model parameters. The model is solved to determine production plans for the product family.
Develop master production schedule (MPS) module
The MPS module makes a production schedule for each product to meet the production plan of the product family (1) in three steps.
Step 1: Design the MPS
Step 2: Make the MPS
Step 3: Control the MPS
Design the MPS
Designing MPS determines scheduling objects, scheduling time, and scheduling parameters. The scheduling objects are products within the product family. Scheduling time is monthly with daily cycles. The scheduling parameters include the number of production days per month (DPM), consumption periods for adjusting MPS in step 2, and time fences for controlling MPS in step 3, as in Table 5.
The MPS parameters.
Parameters
Meaning
DPM (d)
Number of production days per month
BCP (d)
Number of backward consumption periods
FCP (d)
Number of forward consumption periods
DTF (d)
Demand time fences
PTF (d)
Planning time fences
Make the MPS
Making MPS includes the following steps.
Draft an initial production schedule from forecast demand.
Adjust the production schedule according to actual customer orders.
Make the production schedule feasible.
Draft an initial production schedule from the forecast demand. The MPS receives a demand forecast (DF) for each product from the PP module. The DF for each product is derived from the production plan for the product family and the product mix percentage of each product, forecasted by the DM module. The monthly forecasted demand for each product will be divided equally among each day to create an initial production schedule.
Adjust production schedule according to actual customer orders. The MPS also receives actual orders from the DM module. The forecast consumption model (1) will be used to revise the production schedule as customer orders consume the forecast from the production plan. When there are orders that exceed forecast demand, the excess quantity will consume the unconsumed forecast (UF) quantities in the previous cycles (backward consumption) and the following cycles (forward consumption). After consumption is completed, the revised production schedule (RPS) will be defined according to the customer order (CO) and the UF, as follows.
RPS=CO+UF
Make the production schedule feasible. Making the production schedule feasible includes the following steps.
Determine the critical station and its profiles.
Determine the rated capacity (RC) of the station.
Determine the RC of the station.
Check the feasibility of the production schedule.
Adjust when the schedule is not feasible.
The critical station is often the bottleneck station in the production process with the lowest cycle time (CT). The station profiles include parameters of the stations as shown in the following Table 6.
Critical station profile.
Parameters
Meaning
NoM
Number of machine in station
DPP
Number of days per planning period
SPD
Number of shift per day
RHPS (h/s)
Number of regular hours per shift
OHPS (h/s)
Number of overtime hours per shift
DOM
Dates of maintenance
TOM (h)
Time of maintenance
U (%)
Station utilization
E (%)
Station efficiency
CT (sec)
Station cycle time
The rated capacity (RC) of the stations is determined as follows.
AC=NoMDPP*SPD*(RHPS+OHPS-TOM)*E*U*
For the sake of simplicity, with the assumption that the CTs are the same for every product produced in the station, the required capacity is defined according to total demand (TD) per date, and the station CT, as follows.
RC=TDCT*
The production schedule is considered feasible when RC is lower than AC at every scheduling period. If RC exceeds AC at some scheduling periods, loads in these periods will be reallocated to other periods to make the schedule feasible.
Control the MPS
The production schedule must also be controlled by using models of projected available balance (PAB), and ATP, to keep the MPS stable under demand fluctuations (1).
Projected available balance (PAB). Projected available balance (PAB) is used to evaluate the performance of the production schedule and whether scheduled production balances with future demand. With t as the time index of the MPS, the PAB is defined according to the time fences (DTF) and PTF, the beginning on-hand (BOH), the MPS, the CO, and the forecast (F) as follows:
Available to promise (ATP). Available to promise (ATP) is used to promise the customer orders at specific periods in the future. At first, discrete ATP without looking ahead DATP-WOL is defined according to the BOH, the MPS, and the CO before the next MPS. For periods that MPS = 0:
DATP-WOL=0.
For periods that MPS ≠ 0:
t = 1 : DATP − WOL = OHB + MPS − ΣCO
t > 1 : DATP − WOL = MPS − ΣCO
Then discrete ATP with look-ahead DATP-WL is defined according to discrete available to promises - without look-ahead (DATP-WOL). In periods that DATP-WOL is less than 0, the ATP will be set to 0, and the over-promised amount will be compensated by taking the ATP of the nearest previous periods that have positive ATP. Finally, the ATP is defined by cumulating the discrete ATP with a look ahead.
Develop material requirement plan (MRP) module
The MRP module determines the requirement of materials based on the structure of the products and the inventory status of each material to meet the MPS (1). The steps for planning material requirements are as follows:
Determine the production schedule.
Determine the product structure.
Determine the inventory status of materials.
Determine the MRP.
The production schedule is identified from the MPS module. Product structure is often presented through a Bill of Materials (BOM). The inventory status of materials is determined by the inventory status record (ISR), which includes parameters as shown in Table 7.
Inventory Status Record (ISR) parameters.
Parameters
Meaning
LC
Level code
Q
Lot size
LT
Lead time
OH
On hand
SR
Scheduled received
SS
Safety stock
AL
Allocated
Material requirements are planned through variables as shown in Table 8.
MRP variables.
Variables
Meaning
GR
Gross Requirement
SR
Scheduled Receipt
POH
Projected On Hand
PAB
Projected Available Balance
NR
Net Requirement
PORc
Planned Order Receipt
PORl
Planned Order Release
Material requirements are planned over time through the following steps (1).
Determine the gross requirement (GR).
Determine the net requirement (NR).
Determine the planned order receipt (PORc).
Determine the projected on-hand (POH).
Determine the planned order release (PORl).
The GR is determined by the production schedule for the final product and by the PORl of the corresponding parents of the material. The NR, PORc, the POH, and the PORl are determined as follows.
The research methodology will be applied to the case in the following sections.
Demand management (DM)
The DM module used for demand forecasting was included.
DF for product family SS.
DF for individual products in the product family.
Demand forecast for product family
To forecast the demand for the product family of SS in the next 12 months of the next year, 2021, SS demand for the last 3 years, 2018, 2019, and 2020, was collected as shown in Table 9.
Demand of SS for the last 3 years.
Months
1
2
3
4
5
6
7
8
9
10
11
12
2018
2,100
880
550
720
390
1,980
780
1,340
805
1,370
1,305
1,340
2019
2,140
1,220
1,640
2,190
2,420
3,255
3,280
3,550
3,510
3,448
2,250
2,455
2020
2,160
1,320
1,395
1,105
1,410
2,124
1,205
1,460
1,805
1,490
1,310
1,355
Processing the data showed that the data set has no outliers. After analyzing both the trend and seasonal behavior of the data set, the proposed models were auto-regressive integrated moving average (ARIMA), Decomposition, and Winter. Minitab software was used to forecast the three methods. Mean absolute deviation (MAD) and tracking signal (TS) were used as the forecast errors to choose the most suitable model. The forecast errors of the proposed models were shown in Table 10.
Forecast errors of the proposed models.
ARIMA
Decomposition
Winters
MAD
456
591
329
TS
−1.1145
0
0.9018
The Winters method was selected. The forecast result was shown in Table 11.
Demand D of SS for the next year, 2021.
Month
1
2
3
4
5
6
7
8
9
10
11
12
D
1,966
966
1,014
1,151
1,212
2,251
1,546
1,901
1,818
1,873
1,385
1,473
Demand forecast of each product
To determine the DF of each of the six products SS1, … SS6, the demand percentages (DP) of all products were collected in the previous year. The DP of SS1 was forecasted, and the remaining products were done similarly. The DP of SS1 in the last 2 years is shown in Table 12.
Demand percentage (DP) of SS1 in 2019, 2020.
Month
1
2
3
4
5
6
7
8
9
10
11
12
2019
0.004
0.221
0.227
0.006
0.005
0.008
0.007
0.006
0.005
0.082
0.083
0.112
2020
0.231
0.247
0.479
0.068
0.084
0.163
0.314
0.103
0.143
0.201
0.191
0.221
Since the data were collected random and not trending, forecasting methods of Naive Method (NM), Moving Average (MA), and Exponential Smoothing (ES) were proposed. The forecast errors of the proposed models were shown in Table 13.
Forecast errors of the proposed models.
NM
MA
ES
MAD
0.137687
0.0916
0.1031
TS
0.005692
0.00597
−0.00309
The method of MA was selected. The forecast result was shown in Table 14.
The DP for each product in 1/2021.
Model
SS1
SA2
SS3
SA4
SA5
SS6
SUM
DP(1)
0.201481
0.108389
0.084141
0.055543
0.066297
0.484149
1
The demand for each product in 1/2021, D(1), is forecasted by the corresponding DP(1) and PP(1) as shown in Table 15.
The monthly demand of the SS product family D(t), t = 1–12, was forecasted according to Table 11. Data were collected, and model parameters were estimated. Regular and overtime production capacity in units of products per month in the next year are estimated as shown in Table 16.
Production capacity in 2021.
t
1
2
3
4
5
6
7
8
9
10
11
12
RC(t)
1,846
1,182
1,994
1,772
1,846
1,920
1,994
1,920
1,772
1,920
1,920
1,994
OC(t)
692
443
748
665
692
720
748
720
665
720
720
748
Parameters of costs, inventory, and backorder were estimated as shown in Table 17.
Production costs, inventory, and backorder parameters.
Parameters
Value
r(VND/u)
131,011
o(VND/u)
145,384
h(VND/u*m)
22,320
b(VND/u*m)
248,000
MI(u)
2,000
II(u)
0
EI(u)
0
LI(u)
0
MB
0.05
IB(u)
0
LB(u)
0
Applying the Open-Solver add-in in Excel, the production plan in units of products for the next year, 2021, was found in Table 18.
Production plan in 2021.
Month
1
2
3
4
5
6
7
8
9
10
11
12
X
1,846
966
1,014
1,151
1,212
1,920
1,546
1,901
1,772
1,873
1,385
1,473
Y
120
0
0
0
0
331
0
0
46
0
0
0
I
0
0
0
0
0
0
0
0
0
0
0
0
B
0
0
0
0
0
0
0
0
0
0
0
0
With the above production plan of product family SS, the total cost of resources was minimal as follows.
C=2,438,420,206VNDMaster production schedule (MPS)Design the MPS
The production schedule for all models of the product family was made daily for a month. The SS1 would be selected to make its production schedule, and the production schedules of other products were made similarly. The month chosen to be scheduled is January 2021. The MPS parameters were defined in Table 19.
The MPS parameters.
Parameters
Values
DPM (d)
25
BCP (d)
6
FCP (d)
6
DTF (d)
5
PTF (d)
12
January has 31 days, of which 25 are production days, as in Table 20.
31 days in January.
Day
Fri
Sat
Sun
Mon
Tue
Wed
Thu
Fri
Sat
Sun
Mon
Tue
Wed
Thu
Fri
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
Day
Sat
Sun
Mon
Tue
Wed
Thu
Fri
Sat
Sun
Mon
Tue
Wed
Thu
Fri
Sat
Sun
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
The 25 production days of January are shown in Table 21.
The 25 production days in January.
Day
2
4
5
6
7
8
9
11
12
13
14
15
Day
16
18
19
20
21
22
23
25
26
27
28
29
30
Make the MPSDraft the initial production schedule
The DF of SS1 in January was 396, as shown in Table 15. This DF was equally distributed for the production days of the month to make the initial production schedule, as in Table 22.
The initial production schedule of SS1.
Day
2
4
5
6
7
8
9
11
12
13
14
15
DF
15
15
16
16
16
16
16
15
16
16
16
16
Date
16
18
19
20
21
22
23
25
26
27
28
29
30
DF
16
15
16
16
16
16
16
16
16
16
16
16
16
Revise the production schedule according to actual customer orders
Along with the initial MPS, or the DF, were the CO, as shown in line CO in Table 23. The forecast consumption model was used to consume DF by CO to make the revised MPS. The model applied backward forecast consumption first, then forward forecast consumption later, and the number of days for the consumption was 6 days, as in Table 19. The result with the consumed forecast (CF), the UF, and the revised MPS was shown in Table 23.
The MPS of SS1 after consumption.
Day
2
4
5
6
7
8
9
11
12
13
14
15
DF
15
15
16
16
16
16
16
15
16
16
16
16
CO
10
15
10
12
20
12
20
14
12
20
10
10
CF
0
0
0
4
4
4
4
0
4
4
0
0
UF
5
0
6
0
0
0
0
1
0
0
6
6
MPS
15
15
16
12
20
12
20
15
12
20
16
16
Day
16
18
19
20
21
22
23
25
26
27
28
29
30
DF
16
15
16
16
16
16
16
16
16
16
16
16
16
CO
12
12
11
17
10
15
12
20
10
13
10
8
10
CF
0
0
1
1
0
0
4
4
0
0
0
0
0
UF
4
3
4
0
6
1
0
0
6
3
6
8
6
MPS
16
15
15
17
16
16
12
20
16
16
16
16
16
The revised MPS for model SS1 daily in January 2021 was shown in Table 24.
The revised MPS of SS1 in January 2021.
Day
2
4
5
6
7
8
9
11
12
13
14
15
SS1
15
15
16
12
20
12
20
15
12
20
16
16
Day
16
18
19
20
21
22
23
25
26
27
28
29
30
SS1
16
15
15
17
16
16
12
20
16
16
16
16
16
Similarly, the revised MPSs for the remaining models were made as shown in Table 25, with the TD of all models per day.
The revised MPS of all models in the SS family in January 2021.
Day
2
4
5
6
7
8
9
11
12
13
14
15
SS1
15
15
16
12
20
12
20
15
12
20
16
16
SS2
10
6
6
12
9
7
10
8
9
9
9
8
SS3
7
5
6
8
6
5
9
2
3
15
7
7
SS4
7
2
3
8
2
2
12
4
5
3
2
3
SS5
7
3
3
8
2
5
9
2
3
15
2
4
SS6
30
15
35
73
38
38
24
15
34
80
38
38
TD
76
46
69
121
77
69
84
46
66
142
74
76
Day
16
18
19
20
21
22
23
25
26
27
28
29
30
SS1
16
15
15
17
16
16
12
20
16
16
16
16
16
SS2
8
8
9
9
9
8
8
8
9
4
14
7
8
SS3
6
6
7
7
7
6
6
6
7
7
7
7
6
SS4
4
3
11
5
2
2
4
10
4
5
5
4
4
SS5
5
5
3
7
6
5
5
5
5
6
5
5
5
SS6
20
35
35
23
36
90
27
20
30
64
36
38
38
TD
59
72
80
68
76
127
62
69
71
102
83
78
77
Make the MPS feasible
The MPS was checked for feasibility by applying the resource profiles method in (1). The SS production process consists of nine stations: M0, M1, …, and M8. According to station CT, the critical station in the process was M5. The parameters of the M5 station were all estimated and shown in the station profiles shown in Table 26.
M5 station profile.
Station
NoM
DPP
SPD
RHPS (h)
OHPS (h)
DOM
TOM (h)
U (%)
E (5)
ST
CT
M5
1
1
2
8
3
3*
2
94.6
99.8
4
6.5
(*) The 3 days of maintenance were 2nd, 16th, and 30th.
The AC of M5 in minutes per day was calculated as follows.
The AC of M5 in January was shown in the AC lines of the load report in Table 27. For the sake of simplicity, with the assumption that the CTs are the same for every product, the RC in minutes per date was defined according to TD per date and the CT of M5, as follows:
M5 load report in January 2021.
Day
2
4
5
6
7
8
9
11
12
13
14
15
AC
509.6
622.8
622.8
622.8
622.8
622.8
622.8
622.8
622.8
622.8
622.8
622.8
TD
76
46
69
121
77
69
84
46
66
142
74
76
RC
494
299
448.5
786.5
500.5
448.5
546
299
429
923
481
494
OL
−15.6
−323.8
−174.3
163.7
−122.3
−174.3
−76.8
−323.8
−193.8
300.2
−141.8
−128.8
Day
16
18
19
20
21
22
23
25
26
27
28
29
30
AC
509.6
622.8
622.8
622.8
622.8
622.8
622.8
622.8
622.8
622.8
622.8
622.8
509.6
TD
59
72
80
68
76
127
62
69
71
102
83
78
77
RC
383.5
468
520
442
494
825.5
403
448.5
461.5
663
539.5
507
500.5
OL
−126.1
−154.8
−102.8
−180.8
−128.8
202.7
−219.8
−174.3
−161.3
40.2
−83.3
−115.8
−9.1
RC=TDPDCT*=6.5TD*
The required capacity (RC) in January to meet the TD in Table 25 was shown in the RC lines in Table 27. The station load report of the M5 with AC, RC, and overload (OL) was shown in Table 27, where the OL was defined as follows.
OL=RC-AC.
The M5 station was overloaded on days 6, 13, 22, and 27 and under load on the other days, as shown in Figure 3.
The load diagram of M5.
The load on the OL days would be reallocated on the days of underload to make the MPS feasible. The MPS after load reallocation was shown in Table 28.
The revised MPS in January after load reallocation.
Day
2
4
5
6
7
8
9
11
12
13
14
15
SS1
15
27
16
0
20
12
20
35
12
0
32
16
SS2
10
17
6
0
9
16
10
8
10
0
9
8
SS3
7
5
6
6
6
20
9
2
5
0
12
7
SS4
7
2
3
8
2
2
12
6
5
1
2
3
SS5
7
3
3
8
2
5
9
2
3
15
2
4
SS6
30
15
35
73
38
38
24
16
34
79
38
38
TD
76
69
69
95
77
93
84
69
69
95
95
76
Day
16
18
19
20
21
22
23
25
26
27
28
29
30
SS1
16
15
15
17
16
0
12
20
16
16
16
16
16
SS2
16
8
9
9
9
0
12
8
9
0
14
8
8
SS3
6
6
7
8
7
0
8
6
7
5
7
7
6
SS4
5
3
11
5
2
1
4
10
4
5
5
4
4
SS5
6
5
3
7
6
4
6
5
5
5
5
5
5
SS6
20
35
35
23
36
90
27
20
30
64
36
38
38
TD
69
72
80
69
76
95
69
69
71
95
83
78
77
The M5 load report after load reallocation was shown in Table 29 and Figure 4. The OLs of all planning periods were now all negative; the MPS was now feasible.
The M5 load report after load reallocation.
Day
2
4
5
6
7
8
9
11
12
13
14
15
AC
509.6
622.8
622.8
622.8
622.8
622.8
622.8
622.8
622.8
622.8
622.8
622.8
TD
76
69
69
95
77
93
84
69
69
95
95
76
RC
494
448.5
448.5
617.5
500.5
604.5
546
448.5
448.5
617.5
617.5
494
OL
−15.6
−174.3
−174.3
−5.3
−122.3
−18.3
−76.8
−174.3
−174.3
−5.3
−5.3
−128.8
Day
16
18
19
20
21
22
23
25
26
27
28
29
30
AC
509.6
622.8
622.8
622.8
622.8
622.8
622.8
622.8
622.8
622.8
622.8
622.8
509.6
TD
69
72
80
69
76
95
69
69
71
95
83
78
77
RC
448.5
468
520
448.5
494
617.5
448.5
448.5
461.5
617.5
539.5
507
500.5
OL
−61.1
−154.8
−102.8
−174.3
−128.8
−5.3
−174.3
−174.3
−161.3
−5.3
−83.3
−115.8
−9.1
The load diagram of M5 after load reallocation.
Control the MPS
To control the MPS, PAB, and ATP were calculated from the MPS, DF, and CO; the initial OH of 2; the demand time fence (DTF) of 5 days; and the planning time fence (PTF) of 12 days. The PAB and ATP for model SS1 were calculated and shown in Table 30. The MPS of other models could be controlled similarly.
The PAB & available to promise (ATP) for model SS1.
Day
2
4
5
6
7
8
9
11
12
13
14
15
DF
15
15
16
16
16
16
16
15
16
16
16
16
MPS
15
27
16
0
20
12
20
35
12
0
32
16
CO
10
15
10
12
20
12
20
14
12
20
10
10
PAB
0
5
17
23
7
7
3
3
23
19
−1
15
15
ATP
0
5
17
23
11
11
11
11
32
32
12
34
40
Day
16
18
19
20
21
22
23
25
26
27
28
29
30
DF
16
15
16
16
16
16
16
16
16
16
16
16
16
MPS
16
15
15
17
16
0
12
20
16
16
16
16
16
CO
12
12
11
17
10
15
12
20
10
13
10
8
10
PAB
15
15
14
14
14
−6
−6
−6
−6
−6
−6
−6
−6
ATP
44
47
51
51
57
42
42
42
48
51
57
65
71
Material requirement planning (MRP)
Material requirement planning (MRP) determines the materials plan based on the BOM of the products, and the ISR of each material to meet the MPS (1). SS1 was selected as the representative to do the MRP; other products were done similarly.
The MPS of SS1 was shown in Table 28. The BOM of SS1 was shown in Table 31. The materials with units of pieces (P), or rolls (R), included two types of purchasing (P), and manufacturing (M).
BOM of SS1.
Level codes
Materials
Quantity
Units
Types
0
1
2
3
4
5
0
SS1
1
P
M
1
BB
1
R
P
1
HA
1
R
P
1
SS
1
P
M
2
V2
1
P
M
2
DD
1
P
M
3
SBR
1
R
P
3
DG3
1
P
M
3
DB
1
P
M
3
TL
1
R
P
3
V1
1
P
M
4
DG5
2
P
M
4
VM
1
P
M
4
VH3
1
R
P
4
VL
1
R
P
5
VH5
2
R
P
5
VC
1
P
M
The ISR of manufacturing materials was shown in Table 32.
The ISR of manufacturing materials.
Materials
LC
Q
LT (d)
OH
SR
SS
AL
SS1
0
1
0
0
0
0
0
SS
1
1
0
0
0
0
0
V2
2
1
0
0
0
0
0
V1
3
1
0
0
0
0
0
VM
4
1
0
0
0
0
0
DD
2
5
0
0
0
0
0
DG3
3
5
0
0
0
0
0
DB
3
5
0
0
0
0
0
DG5
4
5
0
0
0
0
0
VC
5
20
0
0
0
0
0
The ISRs of purchasing materials were shown in Table 33.
The ISR of purchasing materials.
Materials
LC
Q
LT (d)
OH
SR
SS
AL
BR
2
1
7
50
0
0
0
VH3
2
1
7
50
0
0
0
TL
2
1
7
100
0
0
0
VL
2
1
7
50
0
0
0
VH5
2
1
7
25
125
0
0
BB
1
1
7
0
0
0
0
HA
1
1
7
0
0
0
0
From the BOM and ISR, use the MRP add-ins in Excel to plan the material requirements that meet the MPS. The plan for manufacturing materials was shown in Table 34.
The plan order release for manufacturing materials.
Day
2
4
5
6
7
8
9
11
12
13
14
15
SS1
76
69
69
95
77
93
84
69
69
95
95
76
SS
76
69
69
95
77
93
84
69
69
95
95
76
V2
76
69
69
95
77
93
84
69
69
95
95
76
V1
76
69
69
95
77
93
84
69
69
95
95
76
VM
76
69
69
95
77
93
84
69
69
95
95
76
DD
60
65
70
95
80
90
85
70
70
95
95
75
DG3
30
65
70
95
80
90
85
70
70
95
95
75
DB
30
65
70
95
80
90
85
70
70
95
95
75
DG5
5
135
140
190
155
185
170
135
140
190
190
150
VC
80
80
60
100
80
80
100
60
80
80
100
80
Day
16
18
19
20
21
22
23
25
26
27
28
29
30
SS1
69
72
80
69
76
95
69
69
71
95
83
78
77
SS
69
72
80
69
76
95
69
69
71
95
83
78
77
V2
69
72
80
69
76
95
69
69
71
95
83
78
77
V1
69
72
80
69
76
95
69
69
71
95
83
78
77
VM
69
72
80
69
76
95
69
69
71
95
83
78
77
DD
70
70
80
70
75
95
70
70
70
95
80
80
75
DG3
70
70
80
70
75
95
70
70
70
95
80
80
75
DB
70
70
80
70
75
95
70
70
70
95
80
80
75
DG5
140
145
160
135
155
190
135
140
140
190
165
160
150
VC
60
80
80
60
80
100
60
80
60
100
80
80
80
The plan for purchasing materials was shown in Table 35.
The plan order release for purchasing materials.
Day
2
4
5
6
7
8
9
11
12
13
14
15
BR
84
69
69
95
95
76
69
72
80
69
76
95
VH3
84
69
69
95
95
76
69
72
80
69
76
95
TL
84
69
69
95
95
76
69
72
80
69
76
95
VL
84
69
69
95
95
76
69
72
80
69
76
95
VH5
69
72
80
69
76
95
69
144
160
138
152
190
BB
84
69
69
95
95
76
69
72
80
69
76
95
HA
84
69
69
95
95
76
69
72
80
69
76
95
Day
16
18
19
20
21
22
23
25
26
27
28
29
30
BR
69
69
71
95
83
78
77
0
0
0
0
0
0
VH3
69
69
71
95
83
78
77
0
0
0
0
0
0
TL
69
69
71
95
83
78
77
0
0
0
0
0
0
VL
69
69
71
95
83
78
77
0
0
0
0
0
0
VH5
138
138
142
190
166
156
154
0
0
0
0
0
0
BB
69
69
71
95
83
78
77
0
0
0
0
0
0
HA
69
69
71
95
83
78
77
0
0
0
0
0
0
The plan of manufacturing materials would be the input for CRP. The plan of purchasing materials would be the input for VRP. CRP and VRP were beyond the scope of this study.
Conclusion
The article evaluated the current state of the resource planning system of the company and developed an MRPII system for the company under study with functional modules including DM, PP, MPS, and MRP.
Initially, forecasting models in the DM module were used to forecast the demand of the product family. The PP module developed a production plan that met the demand for the product family and the resource constraints. The MPS module developed production schedules for each product in the product family that matched the production plan. Finally, the MRP module developed material plans that met the production schedules.
The research integrated these functions to reduce production costs and improve customer service levels, then reduce late orders. The article has the following advantages.
Applied forecasting techniques to accurately forecast customer demand.
Applied operations research to minimize the cost of using resources when developing a production plan.
Used the forecast consumption method to adjust the production schedule.
Checked the feasibility and controlled the MPS through PAB and ATP.
Used the ATP model to promise orders efficiently to increase service level.
Established MRP that met the MPS.
However, the research still had some disadvantages. It was only studied and implemented on a product family and just stopped at the planning level. Operations research did not apply to capacity planning. The parameters of the MPS module were chosen by experience. Furthermore, it had not been performed in a practical environment to evaluate the effectiveness of the system. The system would be complete with CRP and VRP. These restrictions lead to further research to improve the system and move towards the ERP system.
Funding
The authors declare that no financial support was received for the research, authorship, and/or publication of this article.
Conflict of interest
The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.
ReferencesNhuPN. Manufacturing Resource Planning MRP II. Vietnam National University Publisher (2020). Available online at: https://www.smashwords.com/books/view/1044194Amad-Uddin KhanMKNoorS. Design & implementation of a bespoke MRPII system for a small and medium enterprise (SME) manufacturing company. J Qual Technol Manag. (2011) VII(I):73–90.NhuPNVoTHTDuyMP. Developing Manufacturing Resource Planning MRP II System. A Case Study in HT Company. (2018). Available online at: www.isem.edu.vnPadhanPC. Use of univariate time series models for forecasting cement productions in India. Int Res J Fin Econ. (2012) 83. Available online at: http://www.internationalresearchjournaloffinanceandeconomics.comCampoEACanoJAGómez-MontoyaRA. Linear programming for aggregate production planning in a textile company. Fib Text East Eur. (2018) 26(5(131)):13–9. 10.5604/01.3001.0012.2525ZhengJ-NChienC-F. Master production schedule and system for excelling enterprise resources (SEER) in the LED industry. IEEE International Conference on Automation Science and Engineering (CASE). (2013). Available online at: https://ieeexplore.ieee.org/document/6654010WeedaPJ. A stochastic model for forecast consumption in master scheduling. Int J Prod Econ. (1994) 35:401–4.NajyRJ. Rough cut capacity planning-RCCP-case study. Adv Theoret Appl Mech. (2014) 7(2):53–66. 10.12988/atam.2014.4612ChenJ-HChenC-T. Using mathematical programming on two-phase order promising process with optimized available-to-promise allocation planning. Int J Comput Int Manag. (2009) 17(3):25–40.WuttipornpunTYenradeePBeullensPvan OudheusdenDL. Finite capacity material requirement planning system for a multi-stage automotive-part assembly factory. ScienceAsia. (2006) 32:307–17. 10.2306/scienceasia1513-1874.2006.32.307NhuPN. Operations Research. Smashwords ebook (2020). Available online at: https://www.smashwords.com/books/view/1052246