Furthermore, a student is not allowed to participate in the practical exams (mock test and presentation) if he/she has more than two absences. Any form belonging to the "table" category must be located in top center alignment with Times New Roman font size 10 point; whereas, the "figure".
Objectives
Literature Review
Basic Concepts and Principles of Forecasting
Module
Time-series Forecasting Methods
- Naïve Approach (Last Period)
- Exponential Smoothing
- Linear Regression (Causal Models)
- Time-Series Decomposition (TSD)
Here is the demand pattern when the available data is suitable for using the moving average method. In the case of a time series model, the independent variable is the time period itself.
Forecast Accuracy
- Procedures
This approach involves decomposing time series data into its fundamental elements, usually consisting of trend, seasonality, and residual (error) components. It is the average of the absolute values of the errors, expressed as percentages of the actual values.
Case Example for Forecasting
Procedure for Moving Average
SUMSQ($firsterror$cell : error cell)/(period-n), where n is the number of periods used in the moving average. You can even try calculating a moving average of two periods to brush up on your Excel skills.
Procedure for Holt’s
Use formula =INTERCEPT(Known y, Known x) for calculating level's value and use =SLOPE(Known y, Known x), for calculating trend value. Next, calculate the trend value using formula =β x (period t level - the previous period level) + ((1-β) x previous period of trend).
Procedure for Time Series Decomposition
Calculate the seasonal 1 (S1) to seasonal 5 (S5) value by averaging the corresponding seasonal value from January 2018 to December 2019. Find the forecast for January 2018 to December 2019 by multiplying the dt bar by the corresponding seasonal (S) value.
Procedure for Winters
Then calculate the trend value using the formula = (β x (level in period t - the level of the previous period)) + ((1-β) x previous trend period). Calculate the seasonal value for the 6th period through the 25th period using the formula =(γ x (1st period sales/1st period level)) + ((1-γ) x S1), then drag the result to the 25th period.
Disaggregate Forecasting
Forecasting Using Minitab
AGGREGATE PLANNING
Tables
- Zero Inventory Plan
- Constant Workforce Plan
- ZERO INVENTORY PLAN (ZIP)
- CONSTANT WORKFORCE PLAN (CWP)
- MIX STRATEGIES: OVERTIME + BACK ORDER
- MIX STRATEGY: OVERTIME + SUBCONTRACT
K = Number of total units produced by a worker in a day K = Average production rate. Average production speed = Number of production Number of working days Table 2.2 Zero stock Total Plan. The cumulative net demand and the number of units produced per worker value is the same as in the ZIP table.
But in the forecasted demand of December 24th, we need to add the expected demand of December 24th with expected ending inventory on December 24th. Number of working days per worker 20 Number of workers at last period (December 23) 60 Ending stock by December 2023 180 Expected Ending stock by December 2024 90. 3 = The maximum overtime per worker per day is 3 hours based on the study case regulation.
ASSEMBLY PROCESS
Assembly Chart
The Product Structure and Bill of Materials (BOM)
- Procedures
Make the next sub-part from Table 3.2, which is sub-2 (SSA-2) by assembling the Motor Mount (EM) and Engine (E) components, with the two components connected by a line to a circle with a diameter of 9 mm form Sub Sub Assembly-2 (SSA-2). Make the entire sub-assembly from Table 3.2 (up to SSA-7) where the two components are connected by a line to a circle with a diameter of 9 mm. Both components are connected by a line to a circle with a diameter of 10 mm and form Sub Assembly-1 (SA-1).
Make the next sub-assembly from Table 3.2 which is Sub Assembly-2 (SA-2) by assembling components Sub Sub Assembly-3 (SSA-3) and Sub Sub Assembly-4 (SSA-4) where the two components are connected by a line to a circle of 10 mm diameter forming Sub Assembly-2 (SA-2). Make the next sub-assembly from Table 3.2 which is Sub Assembly-3 (SA-3) by assembling components Sub Sub Assembly-5 (SSA-5) and Synchronized Elevator (SE) where the two components are connected by a line to a circle with a diameter of 10 mm and shape Sub Assembly-3 (SA-3). Make the next sub-assembly from Table 3.2 which is Sub Assembly-4 (SA-4) by assembling components Sub Sub Assembly-6 (SSA-6) and Stabilizer Bar (SB) where the two components are connected by a line to a circle with a diameter of 10 mm and shape Sub Assembly-4 (SA-4).
Product Structure
All four components are at level 2, which are the materials to make components at level 1. All six components are at level 3, which are the materials to make components at level 2, so write 3 in the level column in it right All six components are at level 4, which are the materials to make level 3 components, so write 4 in the level column on the right.
At level 4, there are Assembly-1 which is made of Sub-Assembly-1 and Skids Landing (LS), Sub-Assembly-2 which is made of Sub-Sub-Assembly-3 and Sub-Assembly-4, and finally Sub-Sub-Assembly-5 which is made of 90° Gearbox (GE) and Tail Rotor (TR). All six components are at level 6, which are the materials to make components at level 5, so write 6 in the level column on the right. At level 6, there is Sub-Assembly-1 which is made of the fuselage (F) and Cabin Doors (CD), and finally Sub-Assembly-2 which is made of the engine (E) and engine mount (EM).
Bill Of Material (BOM)
MASTER PRODUCTION SCHEDULING (MPS)
The value 930 obtained from the above calculation results means that at the end of period 1 the company has about 930 bulldozer toys remaining. The calculation of PAB10 has a condition that t ˃ DTF, then the second formula is used to calculate the PAB value in the tenth period. The value of 930 obtained from the above calculation results means that at the end of period 10 the company has about 930 bulldozer toys remaining.
MATERIAL REQUIREMENT PLANNING (MRP)
As can be seen in Table 5.5, the finished product at level 0 was scheduled for receipt (PORec) in period I = January 2018, so period 1 of lot PORec (S) and component of lot 4 (A-4) must be at level 1 received in the previous period, December 2017. Refer to Table 5.6, PORel Stack (S) and Assembly 4 (A-4) components should be ordered/planned in 1 period before PORec because the lead time of both components is 1. For better understanding look at the train body component (TB) in table 5.6, period 1 PORec TB = August 2017, so for period I PORel TB only by offsetting the period I PORec with preparation time.
As shown in Table 5.6, PORec of the lower-level components is the same as PORel of the upper-level Main Components. Prepare an MRP preparation table for each component according to the format in Table 5.8 below. Then, for period 2 through period 12, the PORec Stack component is populated according to the +1 period/month increment, as shown in Table 5.11 below.
LOT SIZING
- Lot Sizing
- The Silver-Meal Heuristic
- Least Unit Cost
- Part Period Balancing
- Wagner-Whitin Method
- Wagner Whitin
The Part Period Balancing method means that the order horizon is set equal to the number of periods that most closely corresponds to the total holding costs and the setup costs over that period. The results of the calculation in the first period “stop” at the calculation of C(3), so two periods are written on the period row and column “1”, namely December 16 and January 17. Set the order horizon to the number of periods that most closely matches the total holding costs and setup costs over that period.
Next, we transferred the data obtained to the last section of the table, which was filled according to the calculation results. The results of the calculation in the first period "stop" at the calculation of THC 3, so on the period row and column "1" two periods are written, that is Dec-16 and Jan-17. First, fill in the setup cost and holding cost according to the information given, $287 and $195, the same as the previous methods.
INVENTORY CONTROL SUBJECT TO KNOWN DEMAND
- The EOQ Model
- Economic Production Quantity
- History of Kanban
- Kanban Application
- Just-in-Time (JIT)
- Basic Economic Order Quantity Model
- Production Order Quantity Model Example
- Quantity Discount Model Example
- Kanban System
In basic EOQ, it is assumed that the order quantity is received at an instant at an infinite rate, while in EPQ, orders are received at a finite rate over time (H. Mokhtari, 2020). Kanban originated in the Toyota Manufacturing System (Toyota's lean enterprise system) as a tool for managing the flow of production and materials in a "pull" JIT manufacturing process. In a pull system, one process makes more parts only when the other process pulls parts - in effect "pulling" parts from the previous process when needed.
A supermarket is where a customer can get what he needs, when he needs it, in the amount he needs. The downstream process (customer) goes to the upstream process (supermarket) to purchase the required parts (goods) at the required time and in the required quantity. Improved production productivity: Kanban maintains control of the production line, synchronizing all steps in the process.
OPERATION SCHEDULING
Machine Scheduling
- Machine Scheduling for Multiple Jobs in One Machine Sequencing Rules
- Machine Scheduling for Multiple Jobs in Two Machines Johnson Algorithm
- Machine Scheduling for Multiple Jobs in More Than Two Machines Campbell, Dudek and Smith (CDS) Algorithm
Critical ratio (CR): Critical ratio scheduling requires forming the ratio of the processing time of the job, divided by the remaining time until the deadline, and scheduling the job with the largest ratio next. The next smallest processing time is 3, which corresponds to task 4 in column A, so task 4 is scheduled next. The lower path appears to be optimal for this problem with a processing time of 20.
One converts the lower path in Figure 9–10 into a Gantt chart as follows: From time 0 to time 6, Reggie reads A and Bob is inactive. The processing time in the first virtual machine is the processing time on machines 1 and 2 relative to each other. The processing time in the second virtual machine is the amount of processing time of engine m-1 and m to each other.
Employee Scheduling
Note: If two identical two-day periods occur, select the pair with the smallest requirements on adjacent days. In determining the employee schedule, ligo software can be used to calculate the number of workers needed. To get results using Lingo software, it is necessary to use the Deterministic Optimization method.
The limit function is determined based on a combination of the number of working days in 1 week (5 working days) with 2 days off. When creating the employee scheduling table above, it is important to note the value obtained from the Lingo software - see Figure 8.5. Example: From the results obtained by Lingo, X1 has a value of 2, so from Monday to Friday it takes 2 employees for 5 working days with Saturday and Sunday as days off.
BIBLIOGRAPHY