ISL455U-SUPPLY CHAIN MANAGEMENT
Chapter 4: Production Planning in Supply Chains
Introduction
A supply chain’s success strongly depends on providing the right amount of products to customers at the time they are requested. the manufacturer should create a plan for a period, which states what to produce, when to produce and how much to produce in order to achieve the company goals. A plan answering these questions is known as a production plan. This chapter begins by introducing the common concepts and definitions related to production planning.
What is Production Planning?
Production planning can be broadly defined as a management process in manufacturing industries which focuses on the decisions regarding the procurement of components and raw materials and adjustment of workforce level, machines and other necessary resources to make sure that company goals are achieved. In this context, a production plan can be considered as a guide for the company. Hence, a company’s success or failure strongly depends on its production plan. In order to make a production plan, it is necessary to know the demand of the final product.
Forecasting
We can never know the future and that is why we make forecasts. We can classify forecasting methods as subjective ones and objective ones. A subjective forecasting method is based on human judgment. There are several techniques for soliciting opinions for forecasting purposes. Customer surveys, the Delphi method and jury of executive opinion can be examples to these techniques. On the other hand, objective forecasting methods are those in which the forecast is derived from an analysis of data. A time series method is the one that uses only past values of the phenomenon we are predicting. Causal models are the ones that use data from sources other than the series being predicted; that is, there may be other variables with values that are linked in some way to what is being forecasted. In this chapter our focus will be more on time series forecasting methods.
Time series forecasting is a technique which uses the historical data as input and presents data about the future as output. For example, if we want to make a demand forecast, our inputs will be demands in previous years and our output will be demands in next years. There are many forecasting methods which use basic or complicated algorithms. Unfortunately, it is not possible to make a generalization about the success of these methods, their success strongly depends on the data. Hence, selection of the forecasting method which fits the considered data best is crucial.
Based on Nahmias and Olsen (2015) forecasts have the following characteristics.
1. They are almost always going to be wrong. 2. A good forecast also gives some measure of error.
3. Forecasting aggregate units is generally easier than forecasting individual units. 4. Forecasts made further out into the future are less accurate. 5. A forecasting technique should not be used to the exclusion of known information.
There are many forecasting methods. In this chapter, we will study three wellknown and widely used methods: moving average, exponential smoothing and double exponential smoothing with Holt’s method.
Moving Average
Moving average can be considered as one of the simplest and mostly used forecasting methods. In this approach, only N most recent observations are put into account and arithmetic average of these N observations is determined as the forecast value. Note that here N can be an arbitrary value that is determined by the person who makes the forecast.
Double Exponential Smoothing With Holt’s Method
Holt’s method is a forecasting technique that is generally used when the data has a certain trend, such as the values either increase or decrease over time. This method requires two smoothing constants, α and β and has two smoothing equations. The first equation is used for the value of the series and the second equation is used for the trend.
Push and Pull Production Control Systems: MRP and JIT
Different from the push systems, in pull systems, products are moved to the next level only when they are requested. In pull systems, first the orders are received from the customers and then the manufacturing is started based on the customer orders. The earliest of the pull systems is Kanban developed by Toyota, which has exploded into the Just-in-Time (JIT) and lean production movements (Nahmias and Olsen, 2018). Main goal of pull systems is reducing the work-in-process inventory to a predetermined minimum level. In order to achieve this goal, products are moved only when they are requested by the higher level, i.e. there is a demand for that product.
Advantages of JIT and MRP
Nahmias and Olsen (2018) state that both JIT and MRP have some advantages as production planning systems. We can summarize some of the advantages of JIT as follows:
1. Reduce work-in-process inventories, thus decreasing inventory costs and waste, 2. Easy to quickly identify quality problems before large inventories of defective parts are manufactured, 3. When coordinated with a JIT purchasing program, ensures the smooth flow of materials throughout the entire production process.
On the other hand, advantages of MRP can be summarized as follows:
1. The ability to react to changes in demand, since demand forecasts are an integral part of the system (as opposed to JIT which does no look- ahead planning) 2. Allowance for lot sizing at the various levels of the system, thus affording the opportunity to reduce setups and setup costs 3. Planning of production levels at all levels of the firm for several periods into the future, thus affording the firm the opportunity to look ahead to better schedule shifts and adjust workforce levels in the face of changing demand.
The Explosion Calculus
Before introducing the explosion calculus, it may be beneficial to give some basic definitions about the terms that may be faced in an MRP table.
Scheduled receipt: These are items due to be received in a particular time period.
On hand inventory: This is the quantity that is physically present in your warehouse.
Gross requirements: These are the requirements before the netting of on-hand inventory and scheduled receipts.
Net requirements: These are the requirements after netting of on-hand inventory and scheduled receipts.
Lead time: A lead time is the latency between the initiation and completion of a process.
Lot size: Lot size is the quantity of product ordered for delivery in a specific period or manufactured in a single run. In real life, different lot size constraints can be observed.
Product structure tree: It is a hierarchical structure of the raw materials or components that form a certain finished good/final product.
Shortcomings of MRP
Nahmias and Olsen (2018) state that MRP has various shortcomings which can be summarized as follows:
1. Uncertainty: MRP ignores demand uncertainty, supply uncertainty, and internal uncertainties that arise in the manufacturing process. 2. Capacity Planning: Basic MRP does not take capacity constraints into account. 3. Rolling Horizons: MRP is treated as a static system with a fixed horizon of periods. The choice of is arbitrary and can affect the results. 4. Lead Times Dependent on Lot Sizes: In MRP, lead times are assumed fixed, but they clearly depend on the size of the lot required. 5. Quality Problems: Defective items can destroy the linking of the levels in an MRP system.
6. Data Integrity: Real MRP systems are big (perhaps more than 20 levels deep) and the integrity of the data can be a serious problem. 7. Order Pegging: A single component may be used in multiple end items, and each lot must then be pegged to the appropriate item.
Workforce Planning
In this section, we mainly focus on two strategies for capacity and workforce planning. These are chase strategy, which is also known as zero inventory plan, and level strategy, which is also known as constant workforce plan. Under zero inventory plan, the main aim of the company is to manufacture as close as possible to the existing demand pattern.
On the other hand, under constant workforce plan, the main aim of the company is to keep the capacity (i.e. the number of workers) constant during the entire planning period. Hence, once the number of workers is determined in the beginning, no workers are hired or fired during the planning period. Of course, different demand and different capacity needs may be observed in different periods and since no hiring is allowed, the number of existing workers must ensure that no shortages are observed in any period. In addition, in this plan, inventory will be accumulated over time at periods when demand is less than the manufactured quantity and that inventory will be used later on at periods when demand is higher than the manufactured amount.
Note that the two strategies mentioned above are extreme heuristic strategies, one aiming no inventory and one aiming no hiring or firing of workers. Hence, generally neither of them gives the optimal plan. An optimal strategy is generally a combination of these methods and can be obtained by creating a mathematical model which finds a trade-off between conflicting objectives. However, these two strategies can be a beneficial and easy way to find a reasonably good solution to workforce planning problems.