Management Engineering - Industrial Technologies

Completed notes of the course

Complete course

INDUSTRIAL TECHNOLOGIES LESSON 1 - INTRODUCTION TO PRODUCTION SYSTEMS OVERVIEW OF PRODUCTION SYSTEMS The process of new product development consists of five phases: - market analysis , we collect information about the customer needs - product design, we define the product characteristics that satisfy the customer needs - process planning, we determine the sequence of operations and the resources needed to produce the product, but also key parameters (ex: time) - production system d esign, we design the production system based on the process planning - production The supply chain is made of three main processes: procurement , production , distribution . We will focus on the production, which represents the intersection between the new product development process and the supply chain. A production process is a set of activities required to produce goods or services delivered to the market by a company. We will focus on the manufacturing of physical goods. The set of activities is org anized in a specific sequence in order to obtain the final output. The inputs of the production process are production factors while the outputs are products and services. Production factors include operators, equipment, materials, energy, data and infor mation, money. A production system is a subsystem of the company. It uses resources as inputs to provide products and services in order to satisfy the customer needs and the objectives established by the company’s strategy. In short words, t he production system define s where the production process take s place. A production plant is a physical plant where the production system is established. It can be an entire factory or a part of it (the factory can be divided into different production plants). RELEV ANT INFORMATION TO PRODUCTION SYSTEMS Product Part drawing is a drawing used to produce the designed component. It indicates all dimensions, limits and special finishing processes, as well as the material used. Exploded assembly drawing is a drawing tha t shows how the components (identified by numbers) of a product are assembled together . It provides a visual representation of the assembly method . Parts list lists all the components of a product and for each of them it provides different information: the identification number, the name, the drawing reference number, the quantity required, the material, the size, if the component is made internally or purchased. Bill of materials adds some information about the struct ure of the product : the product is structured in levels -> level 0 identifies the final product and each successive level identifies the components that constitute the object of the precedent level. The bill of materials can be used for production but also for maintenance. Process A flowchart is built to provide a graphical representation of the sequence of operations required to obtain a certain product; correspondingly, it shows the material flow for that product. This is typically named as proces s diagram (or similar names). There are different types of flowchart : 1. assembly chart : it shows how and in which order the components are assembled 2. o peration process chart: it adds information about the make or buy option: we use vertical lines to connect components that are produced in -house and horizontal lines to connect components that are purchased . To draw the flowchart you can start from the final output and go backward to identify all the operations necessary to obtain the final output. Route sheet lists in sequence all the operations required to obtain a product and for each operation it provides different information: the identification number, the machine type to perform the operation, the necessary tooling, the set -up time and the ope ration time. Cycle time sheet lists in sequence all the operations required to obtain a product and indicates the total time required for each operation. Flow sheet is a particular method to represent the sequence of operations : the operations are styl ized in icons and ordered in sequence. Production system Layout of a production plant is the representation of the arrangement of machines, work areas and service areas within a production plant. The layout can be drawn in different ways. RELEVANT TER MINOLOGY TO PRODUCTION SYSTEMS Workstation (station, work -center): a collection of one or more machines or a collection of one or more manual stations that perform identical functions. Part: an item that is worked by and moves through the workstations. End item (finished product): an item that is sold directly to customer. Routing: a sequence of workstations passed through by a part. LESSON 2 - FACTORY LAYOUT PLANNING LAYOUT PLANNING: INTRODUCTION Facilities planning determines how tangible fixed a ssets best support achieving the objectives of the activities performed in the facility. In the case of a manufacturing company, facilities planning involves determining how the manufacturing facility (the factory with its own assets) supports production. Facilities planning consists of facility location and facility design. When designing a manufacturing facility, the scope of work includes problems related to: - layout (production areas, production -related/support areas, personnel areas) - material handling system (supporting the required flows) - facility systems (building shell and technical building services) Factory layout planning (FLP) consists in the definition of the physical organization of the factory. FLP aims at finding the most efficie nt location of the departments/shops (areas of activities) within a given building or areas available in a building: - the objective is the optimization of activity relationships between the departments /shops and within the departments/shops - departments /shops might have space requirements, different one from the other - departments/shops should be allocated while respecting the different plant constraints The output of FLP is a technical drawing of the factory layout . There are two types of layout: a) block layout : it is a technical drawing aimed at identifying the location of the departments, including production areas, production -related/support areas and personnel areas. This is a draft layout plan. b) detailed layout: it is a technical drawing tha t enables to specify the exact position of the departments, the structure of aisles, the I/O points, the position of workstation/workplaces within the departments. Typically we draft the block layout and then the detailed layout, it can be an iterative pr ocess. Planning requires to form and locate: - production departments, seen as a collection of machines or manual stations in the production areas; production departments are planned considering the product volume -variety - other areas of activities, s uch as support, administrative and service departments; these areas are planned considering their relationship with product ion departments, and centralizing or decentralizing them. As regards the planning of production departments, we define: - volume a s the number of units produced per product - variety as the number of products produced We can distinguish four cases: - product layout (high volume, low variety): the workstations are located in sequence (line) according to the product working cycle , the line is dedicated to the production of one product ; flow patterns are different depending on the production department . - process layout (low volume, high variety): the workstations that have similar functions are grouped together , every product has i ts own routing that involves passing through some workstations; flow patterns are different depending on the production department. - product family layout (medium volume, medium variety): the workstations are located in a cell, which is dedi cated to the production of a family of products . - fixed location layout (low volume, low variety): the product to be assembled stays in a fixed location while equipment and components are moved as needed. It is used for products of large size and weight. LAYOUT PLA NNING: OBJECTIVES AND PRINCIPLES Planning of material flows requires a hierarchic planning process to effectively deal with all aspects: - top level: effective flow between departments - intermediate level: effective flow within departments - low level: effective flow within workstations Measuring and analysing the activity relationships between the departments is essential to optimize material flows and relationships between the departments . The FLP problem is multi -objective. 1. T he first obj ective function is: min ∑i∑j fij * c ij * d ij - fij = material flow between two areas i,j [units] - cij = unit cost of movement between two areas i,j [€/(unit*meter)] - dij = distance between two areas i,j [meter] The objective function requires to measure the distance, we have three options: - rectilinear distance: we move along the x and y axes - euclidean distance: we move along a straight line - actual distance: we know the actual path to follow 2. The secon d objective function is: m ax ∑i∑j rij * x ij - rij = importance of relationship between two areas i,j (proximity rating) - xij = adjacency of the two areas i,j (it is equal to 1 if they are adjacent , 0 if not) Overall, in developing factory layout altern atives it is important to consider different aspects and subsequent criteria: - layout characteristics - material handling requirements - unit load implied - storage strategies - overall building impact Examples of layout characteristics are: distance travelled, shop floor visibility, occupied and free space. Typical principles, frequently resulting in an effective flow, are: - the minimization of total flow - eliminate flow by eliminating intermediate steps - minimize multiple flows, between two consecutive points/steps - combine flows and operations - the minimization of the costs of the flow - reduce flow by reducing the number of production steps - minimize travel distances with manual handling - eliminate manual handling by mechanizing /automating flow - the maximization of directed and uninterrupted flow paths LAYOUT PLANNING: METHODOLOGY To plan the factory layout w e follow the systemic layout planning methodology (by Muther), which can be divided into two phases: analysis and design. Analysis ^ Analysis of products : the volume -variety information is important to define the layout type to use. The Pareto’s law may either apply or not and this information enables the layout analyst to determine the layout types to use. The Pareto analysis on products support s the strategic definition of factory layout because it leads to a layout product/family oriented or a layout process oriented. ^ Analysis of material flows : the process diagram allows to identify the material f lows between the departments /shops -> based on process diagram s of products , we create the origin/destination matrix of flows : on the vertical axis we put the origin points and on the horizontal axis the destination points (the origin and destination point s are the departments/shops) and in each box given by the crossing of two points we put the material flow between those two points (fij). If we divide the matrix by drawing a line at 45° starting from the upper left box, above the line we have the forward flow and below the line we have the reverse flow. Moreover, the farther a box is from the line, the longer the distance between the two points whose crossing represents that box. We can also create other matri ces: in the boxes we can put fij*cij or fij*wij (w ij = movement effort). For high volume products, the process diagram of the product is directly used to analyse material flows. ^ Analysis of relations between activities : the relationship chart method identifies the relationships between the depart ments/shops. The method leads to define for each couple of departments/shops : - the importance of relationship as a proximity rating (closeness relationship) - the reasons for relationship In case of large volumes and/or big products, the material flow an alysis is more important than the relationship analysis; in case of office layout , the relationship analysis is more important than the material flow analysis. There are intermediate cases in which the two analyses have the same importance. Design ^ defin ition of the relationship diagram + evaluation of alternative solutions : we define the relationship diagram, which is a graph connecting the different departments/shops coded by number ed nodes . ^ definition of the space relationship diagram + evaluation o f alternative solutions: we define the space relationship diagram, which is a graph connecting the different departments/shops coded by numbered shapes ; the size of the shape is function of the area required by the department/shop. The area required is gi ven by the sum of t he area occupied by the equipment and the area to move materials and to let personnel move, which can be multiplied by a coefficient (typically 1,13). We find these information in the departmental service and area requirements sheet. Areas should be typically adjusted and eventually re -located after verifying the available space. It results in alternative block layouts to be further detailed. This process can be supported by automatic algorithms. ^ definition of factory layout + evalu ation of alternative solutions: we define the factory layout considering also other needs and constraints. Also this process can be supported by automatic algorithms, for example CAD factory layout allows us also to visualize the flows between departments/ shops and to assess their intensity/density (the lines representing the flows are of different colours and thickness). Automatic algorithms are formal procedures that can help the layout analyst generate or improve a layout and, at the same time, provide objective criteria to make the evaluation of various layout alternatives that emerge in the algorithmic process. Heuristics are typically embedded in such algorithmic approaches. - CRAFT is one of the first heuristic models and is an improvement procedur e (it improves an existing layout) , it aims at minimizing the moving cost among areas of activities - ALDEP is a construction procedure (it generates a new layout) , it aims at maximizing the proximity requirements among areas of activities A number of commercial packages are available for factory layout planning. Typically, such packages are intended for presentation (electronic drafting tools that facilitate the drawing of a new or an extent layout) or as a layout evaluation tool ( able to calculate spe cific performance indicators required for the evaluation). LESSON 3 - JOB SHOP JOB SHOP: GENERAL FEATURES We can classify the production plants in process plants and manufacturing plants (we perform the parts production and the assembly). We can class ify the production systems based on the configuration of the plant and w e will consider only manufacturing plants. Now we focus on job shops (process layout), adopted to produce many low volume products. Job shop production is characterized by high flexi bility (high variety) and low efficiency (low volume). In a job shop, machines are grouped on the basis of the homogeneity of the production technologies they provide (technological processes they support) ; this means that machines are organized according to the operations they are capable to perform. The groups of machines are called functional departments . In a job shop, each product has its own routing in the system and production flows are jumbled, as intersections of flow paths of different products happen often. Each product can have a preferable routing and one or more alternative routings (more costly). ! In parts production, product and part are terms used as synonymous. In a job shop, materials are moved according to the required product routings (from one department to another). The logistics (material handling through the production system) is characterized by high flexibility; this results from the presence of transpor ters and storages within the system. In a job shop, the labour is divided in departments according to task specialization: workers assigned to a department are skilled in the production technology performed in that department. JOB SHOP: STRENGHTS AND WE AKNESSES The main strength of job shop is the high flexibility: - short to medium term flexibility - mix: ability to meet market requirements in terms of product mix variation - volume: ability to meet market requirements in terms of volume variation - product (customization): ability to meet market requirements in terms of customization - medium to long term flexibility - product (innovation) : ability to start new productions - expansion : ability to increase the production capacity by introdu cing brand new machines Thanks to basic flexibility in the production system: - machine flexibility: each machine can ideally work on different products - material handling flexibility: the material handling equipment can be scaled up /down and moved - routing flexibility: we have the possibility to follow alternative routings Consequently, we have: - low impact of breakdowns, thanks to machine and routing flexibility - low obsolescence of the system, thanks to medium -long term flexibility The main weaknesses of job shop are: - limitations in (machine) efficiency - set -up times, reducing the available time - quality losses due to alternative routings, reducing the running time - speed losses due to alternative routing s, reducing the running t ime - qualit ative characteristics of the product can vary for different pieces depending on the machine used - production management is difficult - high WIP , to avoid machines waiting for material starvation - lead times are long (because of queues) an d characterized by high variability - difficulties in estimating delivery lead times - low utilization rate of machines , due to material starvation - it is difficult to calculate the production capacity of a job shop, that depends on: - mix of jobs that have to be manufactured - complexity of pieces to be manufactured - technological characteristics of jobs - possibilit y to use alternative routings - definition of the lot sizing - number of machines and their state - ability to schedule jobs JOB SHOP: SYSTEM DESIGN Typical questions related to system design are: - how many machines do we need to meet demand? - how many operators do we need to meet demand? - where are the bottlenecks? - what happens if the production mix changes? - what is the effect of reducing setup times or lot sizes? - what happens when a machine breakdown happens? - what is the effect of adding another machine to the system? We can roughly design a job shop system by following eight steps: 1. definition of the production mix - identify all the product types - estimate the yearly demand for each product type - define the lot size for each product type 2. definition of the technological cycles - define the main process to produce each product type - if po ssible, define alternative processes ( -> alternative routings) 3. identification of the machine types - on the basis of the technological cycles, it is possible to identify all the machine types that are necessary to manufacture the production mix 4. ca lculation of the total time of operations - for each product type, calculate the total time of the operations that have to be performed on the same type of machine (T ij with i = index of the machine type and j = index of the product type) 5. calculation o f the yearly workload NH i for each type of machine i Qj = quantity of product type j to be produced [pieces/year] SR ij = scrap rate ( = quantity with defects/total quantity produced) STT ij = setup time, NL j = number of setups = number of lots (= quant ity /lot size) Ai = availability of the machine type i , HC i = human coefficient, TR i = trial rate The yearly workload is computed as a gross value (it includes detractors such as time with defective parts, setup times, machine downtimes, time lost due to operator unavailabilities, time spent for trial production). 6. calculation of the number of hours available AH i for each machine type i AH i(s) = WH i(s) * SE WH i(s) = yearly working time available (depending on the number of shifts per day) = number of hours per shift * number of shifts per day * number of days per year SE = scheduling efficiency (0 ≤ SE ≤ 1) 7. calculation of the number of machines of type i necessary to manufacture the production mix, given the yearly demand: NM i(s) = NH i/AH i(s) The higher the number of shifts : - the lower the number of necessary machines - the higher the number of necessary operators 8. evalua tion of the number of shifts/day, computing the yearly costs adopting 1, 2 or 3 shifts/day WF i(s) + OC i(s) + NM i(s)*CA i*m i + FC i(s)*f i WF i(s) = yearly cost of direct and ind irect labour OC i(s) = yearly operating costs NM i(s) = number of type i machines CA i = cost of a type i machine FC i(s) = installation costs of facilities mi - fi = coefficients used to split costs on the machine lifetime - facility lifetime The number of machines is obtained by approximation (rounding up or down), this depends on: - machine type cost - possibility to outsource the production of some product types - possibility to adopt alternative processes for some product types The utilization of machines can be calculated as: UT i = NM i/integer value of NM i We have alternatives lead ing to trade -offs: - optimal shift for each machine type / department - minimization of yearly costs for the department - extra costs: higher WIP and required space, costs related to shared resources - same shift for all the machine types / department s - yearly costs for the departments not minimized LESSON 4 - MANUFACTURING CELLS MANUFACTURING CELLS: GENERAL FEATURES Now w e focus on manufacturing cells (product family layout), adopted to produce some medium volume products. Cellular m anufacturing is characterized by mid flexibility (mid variety , lower than in job shop ) and mid efficiency (mid volume , higher than in job shop ). When cellular manufacturing is applied, parts are grouped into part families and machines into cells. Differe nt types of production technology (technological processes) are involved in the same cell. The parts are grouped on the basis o f their similarity in terms of production flow. The machines are grouped on the basis of the processing requirements of the part families. Every cell is dedicated to a part family and each part has its own routing within the cell (this is the case when no inter -cell move is required -> case of complete cell independence). Applying the cellular manufacturing typically leads to re -arrange the existent equipment on the factory floor. A typical question required to then make the system design is “which machines and their associated parts should be grouped t ogether to form cells?”. As a system design result, the layout is then changed in order to the re -arrange the physical organization of the resources on the factory floor (re -layout). As re -layout plan, a U -shaped flow line can result from the concept of cellular manufacturing. This allows more efficient use of the workers tending the machines, while also promoting a better communication (in case of more workers involved in the cell). Moreover, the worker has more visibility of the entire cell. A particul ar case is the virtual manufacturing cell (VMC): the machines that belong to a VMC are not physically located together, but are identified as a “virtual” group only by means of the production planning and control system. MANUFACTURING CELLS: STRENGTHS AN D WEAKNESSES The main strengths of manufacturing cells are: - rationalization of material flows, based on the similarity of parts and their processing requirements - standardization (the cell is dedicated to a product family) - standard system to hold and clamp the workpiece - availability of common tools for the part family - setup time reduction , part families often have setup sharing potential - production management is easier (lower variability, lower WIP, shorter lead time) - job enlargement + job enrichment for employees - team work within the cell - unification of product and process responsibilities , the workers of a cell are responsible for the processes required by the product family assigned to the cell and therefore for that product fami ly - more control on the qualit ative characteristics of the products Overall, compared to the job shop: - WIP reduction - lead time reduction (shorter queues and reduced setup time) also considering variability - more reliable estimates of delivery lea d times The main weaknesses of manufacturing cells are: - problems related to production mix variabilit y - difficulties to balance the workload between cells : each cell is dedicated to a product family - problems related to breakdowns ( the number of machines of the same type is lower than in job shop) - difficulties to manage technological operations outside the cells (a machine can be kept as a common resource outside the cells -> the material flows are intertwined) - difficulties with the applicati on to all stages of the production chain - in some cases, necessity of more machines than in a job shop (ex. different families need the same machine type but they are assigned to different cells -> duplication of machines) GROUP TECHNOLOGY: BENEFITS, STEPS AND METHODS Group technology (GT) is a methodology aimed at capitalizing on underlying similarities of parts and their required activities/operations. Cellular manufacturing is a result of the application of the GT, with different benefits. Nevertheless, it is worth to remark that GT brings benefits in different areas, such as design and engineering (other benefits closely related to production and manufacturing cells have been already discussed). Group technology is a m ethodology developed in six steps: 1. data collection regarding production mix and technological routings 2. classification of products 3. standardisation of products 4. standardisation of product routings 5. identification of product families 6. identifi cation of machine groups forming the cells The identification of the product families can be based on : - geometrical features - technological features We can follow three different methods to identify the product families and the machine groups: - inf ormal method: it relies on the opinion of experts - part coding analysis: the parts are coded and then proper filters are applied to the parts to create classes of parts based on some similarity of their design attributes - production flow analysis (PFA) : we will use this method, which consists of three steps: 1. clustering analysis , which can be performed in two ways: - rank order clustering (ROC) : we build a matrix where we have the machine types on the rows and the parts on the columns; at each inter section between a machine type and a part we put 1 or 0 . 1. read each row as a binary number ( with j = 1 to J, binary number of the machine type i = ∑j aij * 2 J-j ) 2. order rows according to descending binary numbers 3. read each column as a binary number (with i = 1 to I, binary number of the part j = ∑i aij * 2 I-i ) 4. order columns according to descending binary numbers 5. if on steps 2 and 4 no reordering happened, go to step 6; otherwise go to step 1 6. stop At the end, we identify the parts that have to be produced by the same machine types in order to create the cells. There can be exceptional parts that need a processing out of the main cell -> possible solutions are: inter -cell movements, duplication of ma chines, alternative routings, buy operations from third parties. Example: slide 40 -46 - similarity coefficients: we build a matrix where we have the machines on the rows and the parts on the columns; at each intersection between a machine and a part we p ut 1 or 0. 1. compute the similarity coefficeint betwee n machines i and j: Sij = a ij/(a ij + b i + c j) with a ij = number of parts worked by both machines , bi = number of parts worked by only machine i cj = number of parts worked by only machine j 2. fulfil the similarity matrix with S ij 3. build the hierarchical tree (dendogram) 4. given a threshold, group machines with higher similarity coefficient -> the groups of machines form the cells Example: slide 49 -62 2. graph partitioning 3. mathematic al programming MANUFACTURING CELLS: SYSTEM DESIGN The successful implementation of manufacturing cells requires addressing different problems: - (first issue) the selection of machine types and part family for a cell - the cell design (layout, producti on and material handling requirements) - the cell operation and control (production lot size, scheduling, number and type of operators, type of production control and methods to measure the performance of the cell) To rough design the cell, the same appro ach used for the job shop can be adopted: - calculate the number of machines of type i necessary in the cell - evaluate the number of shifts per day, computing the yearly costs adopting 1, 2 or 3 shifts per day LESSON 5 - TRANSFER LINES TRANSFER LINES : GENERAL FEATURES Now we focus on transfer lines (product layout), adopted to produce few high volume products. Transfer line production is characterized by low flexibility (low variety) and high efficiency (high volume). Each transfer line consists of a series of machines where a single product type flo ws or a limited number of similar product types flow. The transfer line is integrated with the material handling system , which moves the pieces from the upstream machine to the downstream machine. Typic al features to describe a transfer line are: - high and stable demand of products - great emphasis on the efficiency - fixed routing through the machines - sequential production flow - serial dependency (flow through a series of machines) - highly autom ated system The transfer lines can be classified according to: - the specific layout : - linear, the machines are placed in a line - u-shaped, the machines are placed in a u -shape - rotar y, a round table turns and places the pieces in front of the machines (circular transfer) - the way the material handling system works : - paced, the movement of the piece from one station to another is synchronous for all the stations - unpace d, the movement of the piece from one station to another is asynchronous ; there may be buffers between the stations - the production management mix : - single -model, the transfer line produces only one product type - multi -model , the transfer line produce s more product types alternatively (we have setups) In both cases we ha ve to define the batch size for each product type; in the multi -model we have also to define the sequence of the product types to produce, which can depend on setup times. TRANSFER LINES: STRENGHTS AND WEAKNESSES The main strengths of transfer lines are: - simple production management (only batch sizing and batch sequencing in case of multi -model) - high machine utilization , due to the low setup time and stable demand of products - low space occupied : machine are placed in line -> a transfer line is a compact system Overall, subsequent performances are : - low WIP : the production is serial and stable - low lead time (no/short queues, low setup time, quick handling) also considering variability - low need for workforce , due to the high automation - qualitative characteristics of products are stable (all pieces are processed by the same machines) The main weaknesses of transfer lines are: - low flexibility : designed to produce a limited set of products + batch production - high investment needed (high automation for workstations and material handling system) - long time required to start new productions (need to reconfigure the line) - high risk of obsolescence (long time to start new productions, lifetime related to products’ lifetime) - significant impact of failures : if one machine breaks, the line stops (solutions: buffers, parallel machines) TRANSFER LINES: SYSTEM DESIGN A remarkable similitude exists between lines in manufacturing, both for parts production and assembly. Most t echniques (as balancing and buffering) can be used, adapting their application to the specific challenges of the line. We will focus only on line balancing, the discussion on line buffering is postponed to the assembly. The system design typically starts with the definition of technological cycle/routing of the product ty pe(s): 1. definition of the production process (also named fabrication cycle) - the process is composed by elementary operations (task s) organised in a certain sequence - a precedence gr aph/diagram (or equivalent chart) is defined to model the sequence (some constraints (precedences) between operations exist and are expressed in the graph/diagram) 2. definition of the (deterministic) time for each operation and calculation of the total w ork content of the production process (which is equal to the sum of the operation times within the process) The system design typically continues with the identification of all the machine types required by the production process and the line balancing ba sed on a given cycle time. The main objective is to achieve the target production volume of the line, considering: - the output of the line per unit time, that is the throughput rate or a given production capacity - the cycle time (as the time interval between the exit of one piece and the exit of the next piece from the line) used to drive the line balancing The cycle time is: - usually expressed in seconds/unit (or minutes/unit) in these manufacturing systems - the reciprocal of the production capacity required by the line, being the production capacity computed as the required production volume to be processed in a given time (it’s a target) (ex. cycle time = 2 minutes/piece -> production capacity = 30 pieces/h our) - the longest processing time among the workstations within the line ( it’s obtained from the line design) We compute the minimum number of stations required within the line to achieve the given cycle time: Nmin = total work content/cycle time (CT) (it must be rounded up) Effects of good or bad line balancing can be equivalently measured as throughput (rate) : the workstation with the longest processing time = the workstation with the lowest throughput rate. Line balancing can be done by means of an iterative procedure that helps generating progressively the workstations and then allocating the operations to the workstations. Using the precedence diagram and the operation times (T k), the operations are al located to the individual stations so that the sum of operations at each station doesn’t exceed the workstation time T s (≤ CT to meet the market requirements). We can choose between different methods to allocate the operations, such as: - method MaxDu r: the operations are ranked in order of decreasing processing time ; the operation with the longest processing time is allocated as soon as possible (sub -criteria: alphabetic order) - method MaxNfol: the operations are ranked in order of decreasing number of immediately following operations; the operation with the highest number of immediately following operations is allocated as soon as possible (sub -criteria: MaxDur) - method RPW (ranked positional weighting) : the operations are ranked in order of decrea sing positional weight (PW i = time to perform operation i + time to perform all successor operations to operation i) ; the operation with the highest PW is allocated as soon as possible (sub -criteria: MaxDur) We can choose between two approaches to allocat e the operations: - task oriented approach (TOP) : when the remaining time of the workstation is not sufficient to assign the i-th operation, a new workstation is opened - station oriented approach (TOP) : when the remaining time of the workstation is not sufficient to assign the i -th operation, before opening a new workstation, other available operations are considered to fill -up To balance the line we can also divide the slowest station into two sub -stations positioned in parallel. To compute th e efficiency of the design ed solution, we can use performance indicators such as line efficiency: Line efficiency (E) = N min /N = total work content/total time provided, where total time provided = N*CT N = actual number of stations, it has to be greater than or equal to N min . It is not always possible to have a perfectly balanced solution in which all stations are highly utilised. We have perfect line balancing when all workstations within the line have the same processing time. Rough design of a transfer line: single -model 1. define the technological cycle and operations of the product type 2. identify all the machine types that are needed and balance the line on the given CT [s/piece] 3. compute the theoretical production capac ity: TPC = 3600/CT [pieces/h] 4. compute the actual production capacity: APC = TPC * A * (1 -SR) [pieces/h] where A = line availability (0 < A ≤ 1), SR = scrap rate (0 ≤ SR < 1) 5. compare the actual production capacity and the demand; if necessary, modify the line and go to step 2 Rough design of a transfer line: multi -model We assume that: pieces are manufactured in batches ; changing production from one batch to another requires a setup; setup times do not depend on the production sequence. 1. identify the production mix 2. define the technological cycle and operations of the product types (in the production mix) 3. identify all the machine types that are needed and balance the line for each product type 4. compute the cycle time for each product type j : CT j = max h {TL jh} [s/piece] where TL jh = unit working time of product type j at workstation h 5. com pute the whole time to produce a batch of product type j: T j = CT j * H + CT j * (Q j - 1) + STT j where H = number of workstations in the line, Q j = batch quantity of product type j [pieces/batch], STT j = setup time related to a batch of product type j [s/batch] -> approximating: T j = CT j * Q j + STT j [s/batch] 6. compute the time needed for a set of batches (within a production campaign): T = ∑j=1N Tj where N = number of batches in the campaign 7. compute the average theoretical production capacity: TPC = 3600* ∑j=1N Qj /T [pieces/h] 8. compute the actual production capacity: APC = TPC * A * (1 -SR) [pieces/h] 9. compare the actual production capacit y and the demand; if necessary, modify the line and go to step 3 LESSON 6 - ASSEMBLY SYSTEMS MANUAL ASSEMBLY SYSTEMS: GENERAL FEATURES Manual assembly systems are composed of: - workstations, each station has one or more operators: the elementary tasks are assigned to the operators in case of manual assembly process; manual assembly tasks do not usually require expensive tools - material handling systems to move the components, the subassemblies and the assemblies; there are different types of mate rial handling system: belt, roller conveyor, overhead conveyor, forklift trucks, AGV Material feeding to the workstations may consist of different methods, such as: - supplying small stocks of components to each workstation (for small and cheap pieces) - bringing large and heavy components/subassemblies to the workstations synchronously with the main assembly on progress - using assembly kits (set of components, subassemblies and specific tools) The supplying of components can be sequential or continuous. We can identify four types of assembly system, based on the product volume -variety to be assembled: - fixed position assembly: big/heavy and complex products - assembly shop: low volume products, high variety of products - assembly cell: mid volume prod ucts, mid variety of products - assembly line: high volume products, low variety of products FIXED POSITION ASSEMBLY In fixed position assembly, the product is assemble d in a single site rather than being moved through a set of assembly stations. Materials, equipment and tools are brought to the site. It is used to assemble products of high volume or weight ; typically these products are also very complex. This assembly system requires skilled operators. The assembly process is typically long and therefore e ach site is characterized by a low production capacity . ASSEMBLY SHOP An assembly shop consists of different stations and to each station is assigned a phase of th e assembly process of the product type; more stations are then visited according to the sequence of phases to be performed to complete the entire assembly process of the product. The same phase can be assigned to more stations -> this assembly system is f lexible. It is used to assemble many low volume products. ASSEMBLY CELL In assembly cells, the products move during the assembly through stations with some flexibility in their routing (with slightly different flows for different products). It is used to assemble families of products or similar products. In manual assembly cells, the organization of work is based on teamwork, following rules and organizational solutions similar to those adopted for manufacturing cells. Cell responsibilities can be assigned to operators. The operator can perform all the tasks of the assembly process allocated to the cell. It is easy to assign testing and adjusting (and repairing) of the assembled product to the product. LESSON 7 - ASSEMBLY LINES ASSEMBLY LINES: GENERAL FEATURES Each assembly line consists of a series of stations where the product is progressively assembled. The components are moved to the stations. It is used to assemble few high volume products. The assembly lines can be classified acc ording to: - the specific layout: - straight -line flow, the machines are placed in line - U/S/W -shaped flow, the machines are placed in U/S/W -shape - O-shaped flow, the machines are placed in circle - the way the material handling system works: - pace d, the movement of the piece from one station to another is synchronous for all the stations - unpaced, the movement of the piece from one station to another is asynchronous ; there may be buffers between the stations - the production management mix - single -model, the line assembles only one product type - multi -model, the line assembles more product types alternatively (batches) - mixed -model, the line assembles more product types alternatively (pieces) In the multi -model and in the mixed -model we ha ve to define the sequence of the product types to assemble , which can depend on setup times. Depending on the way the material handling system works, assembly lines can be paced or unpaced. In a paced line, a common cycle time is given and restricts processing times at all stations . A synchronous movement of the pieces through the line occurs. The pace is either kept by: - an intermittent transport : the system moves the piece from one station to another at defined intervals; the piece stops at every station, but is automatically transferred as soon as a given time span is elapsed - a continuously advancing material handling device : the system on which the piece is located moves continuously; it forces the operato r to finish his task before the piece has reached the end of the station The intermittent lines can be paced by machine or by operators: - machine -paced lines: the movement of pieces is paced by a timer and given by the cycle time of the line. Cycle time and production capacity are perfect ly controlled, but there is the problem of unfinished pieces (there is the risk that the operator fails to complete his task) . - operator -paced lines: the movement of pieces is paced by operators -> the system moves only after all operators have given the ir approval. There is no problem of unfinished pieces, but the cycle time is variable and it is determined by the slowest operator. In continuous flow paced lines, the material handling system moves at a constant speed and operators follow the piece on wh ich they have to perform the assembly tasks (or they move with it on a platform). The distance between two pieces is lower than the length of the station , so that the operator has some extra time to complete his work on the piece i f necessary (the systema tic use of extra time is a problem) . We can distinguish two cases based on the possibility for operators to stop the line: - case 1, operators can’t stop the line: cycle time and production capacity are perfectly controlled and cycle time can be exceeded occasionally, but there is the problem of unfinished pieces. - case 2, operators can stop the line: there is no problem of unfinished pieces and cycle time can be exceeded occasionally, but cycle time and production capacity are not perfectly controlled. In a n unpaced line, each station has a different processing time and buffers are put between consecutive stations to absorb this variability. An asynchronous movement of the pieces through the line occurs. Blocking: if the upstream station is quicker than the downstream station, the number of pieces in the buffer increases until it is fully saturated; at this point the upstream station has to stop. Starvation: if the downstream station is quicker than the u pstream station, at a certain point t he downstream station has to stop because there aren’t pieces to work on. Blocking and starvation issues also depend on the size of the buffers: the higher the buffer size, the lower the occurrence of these issues. Th ere is no problem of unfinished pieces and cycle time can be exceeded occasionally, but cycle time and production capacity are not perfectly controlled. ASSEMBLY LINES: STRENGHTS AND WEAKNESSES The main strengths of assembly lines are: - rationalizatio n of material flows (the product moves along the line and each workstation is fed with its pertinent components) - low WIP : the production is serial and stable - limited space requirement : machines are placed in line, an assembly line is a compact system - labour training might be easy (high repetitiveness, few tasks) - low cost for workforc e (easy and repetitive tasks) The main weaknesses of assembly lines are: - low flexibility : designed to assemble few products + batch production - long time required to start new producti ons (need to reconfigure /rebalance the line) - repetitive work (solution: operators can be moved from line to line everyday) - line balancing might be difficult , especially for complex products and/or with large number of var iants ASSEMBLY LINES: SYSTEM DESIGN Initial considerations for design (valid for all lines): - evaluation of the time of each assembly operation - computation of the cycle time - definition of other balancing constraints - definition of the balancing o bjectives The design consists of assembly line balancing (ALB), which means assigning the operations to the stations so that the time required at each station is approximately the same. The focus is on manual assembly lines. There may be some constraints: - cycle time - precedence relationships among operations - incompatibility between operations that cannot be assigned to the same station - opportunity or necessity to assign some operations to the same station - constraints related to space - constraint s related to workers - constraints related to the material feeding The objectives of the assembly line balancing can be: - minimizing number of workstations (to decrease costs) - minimizing cycle time (to increase the production capacity and therefore p rofits) - maximizing efficiency - minimizing production costs - maximizing profits The technical objectives are: - minimizing the number of stations, given the cycle time - minimizing the cycle time, given the number of stations - minimizing the total idle time: IT = n * CT - ∑i=1N ti where: n = number of stations, CT = cycle time, N = number of operations, t i = time to perform operation i ! minimizing the number of stations, given the cycle time, is equivalent to minimizing the total idle time. - minimizing the probability of no completion - in a machine -paced line - in a continuous flow line, in case the operator can’t stop the line - minimizing the probability that the time of operation in one or more stations exceeds the cycle time - in a n operator -paced line - in a continuous flow line, in case the operator can stop the line The economic objective is minimizing the total expect ed cost: TEC = LC + ECUT where: LC = line cost (equipment + operators), ECUT = expected cost of unfinished operations If we increase the number of stations, LC increases and ECUT decreases -> we must manage the trade -off. Depending on the type of line a nd specific constraints, different technical objectives are considered. Techniques as balancing and buffering are used, adapting their application to specific challenges of the line. As in manufacturing systems, line balancing can be done by means of an i terative procedure that helps generating progressively the workstations and then allocating the operations to the workstations. Using the precedence diagram and the operation times, the operations are allocated to the individual stations so that the sum of operations at each station doesn’t exceed the workstation time . For machine -paced lines, we refer to the concept of capacity buffering: we keep some idle time to reduce the risk of no completion -> limit of utilization rate < 100%, equivalent to capa city bufering First m ethod: utilization rate (machine -paced line) The utilization rate of an operator (station) is defined as: UR = ∑iϵS ti / CT where t i = mean time to perform task i, S = set of tasks assigned to the operator, CT = cycle time For each operator, the following constraint has to be verified: UR ≤   is the maximum value of the utilization rate (0 <  ≤ 1). Typically, the value of the utilization rate increases as we move along the line: it has the lowest value at the first stat ion and the highest value at the last station because the impact of no completing the first task is greater than the impact of no completing the last task. The steps to follow are: 1. draw the precedence graph 2. calculate the total task time T (sum of all task times) It can be defined as the total assembly work content or total assembly time. 3. calculate the cycle time: CT = available time/demand 4. calculate the minimum number of stations: K* = T/(CT* ) 5. assign operations to stations, respecting the constraints The maximum utilization rate is adopted in order to evaluate the allocation of the operations to the stations. If there is more than on e operation available to be assigned, use a rule to prioritise operations. Second m ethod: probability of no completion (machine -paced line) In case of probability of no completion adopted as the criteria in the iterative procedure, for each task the following constraint has to be satisfied: P k ≤ P* where P k = probability of no completion of task k, P* = maximum probability of no completion The steps to follow are: 1. calculate the remaining time related to task k: RT k = CT - ∑iϵS ti where S = set of tasks assigned to the operator (station), the new task k to be assigned is included 2. calculate the variable associated to the remaining time: Z k = RT k/square( ∑iϵS i2) where i= standard deviation of the time required to perform task i 3. calculate the probability of completion: find (Zk) from normal standa rd distribution table 4. calculate the probability of no completion: P k = 1 - (Zk) Demonstration: slide 53 Limiting the utilization rate leads to a limitation of the probability of no completion. Increasing the remaining time leads to increased probabil ity of completion/decrease d probability of no completion. -> the two methods have the same objective, which is limiting the probability of no completion. There may be a fixed assembly station where all the incomplete pieces are brought; the operator of this station has the skill to work on all the pieces. For unpaced lines, we refer to the concept of inventory buffering: we use buffers to absorb the variability of assembly times -> limit of utilization rate = 100%, complemented with inventory buffering The basic criteria for allocating operations is to maximize the utilization of the station time. The production capacity depends on the bottleneck station, which d ictates the cycle time of the line. The production capacity also depends on blocking and starvation , that can be reduced thanks to buffers. Buffer size and CV (coefficient of variation of the assembly times) determine the production capacity: the higher the buffer size, the lower the probability of blocking and starvation and the higher the production capacity . The marginal benefit of increasing the buffer size is decreasing. Buffer size can have important influence: - buffers enable to limit the reduc tion of production of the line, due to assembly time variability - buffers lead to higher investment, space requirement, WIP Therefore, it is crucial to find the optimal buffer size. Moreover, the importance of buffers increases with the CV: given the t arget production capacity, if the CV increases, you have to increase the buffer size to reach the target production capacity. -> Line balancing and inventory buffering are two techniques used together to optimize the design of unpaced lines. The optimal b uffer size is influenced by the goodness of assembly line balancing. We need to protect more the stations with higher times, so we put bigger buffers near them. For continuous flow paced line, in case operators can’t stop the line, we refer to the concept of time buffering: we add extra length to the conveyor so that the operator has extra time to eventually complete his task -> limit of utilization rate = 100%, c omplemented with time buffering. The station length sizing is equivalent to insert ing a time buffer to protect the line from assembly time variability, thus limiting the problem of unfinished pieces (no completion). L ≥ D, where D = CT * V = V / PC with CT = cycle time, V = conveyor speed, PC = production capacity When we define the length of the station (L) , we have to consider the trade -off: - the higher the length, the lower the probability of no completion - the higher the length, the higher the sp ace required and the higher the flow time (FT = L/V) Another solution to the problem of no completion is to add two shared zones for each station: an upstream zone and a downstream zone where the operator can complete his task. These zones are called open stations. There is the risk of interference between two operators who want to work in the same zone. MULTI -MODEL LINES General features of multi -model lines: - product types are made in batches - model variation can be wid e (different product types wi th variations ) In multi -model lines: - we determine the batch size and sequence in which the product types/models have to be launched onto the line to optimize the setup times between batches (the higher the setup time, the higher the batch size) - we k eep high inventories to have products in stock during the periods when other products are assembled - we determine the number of stations and operators in the line, while dealing with the different requirements of the product types, so we determine the li ne balancing for each product type with distinct features and requirements The steps to follow to balance a multi -model line are: 1. compute the minimum number of stations of the line: K* = ∑j (Q j * ∑iϵSj Tij) / H where Q j = quantity of product type j (yearly demand) , S j = set of tasks related to product type j, T ij = mean time of task i of product type j, H = number of available hours, = maximum value of the utilization rate 2. compute the (target) cycle time for each pr oduct type j: CT j = ∑iϵSj Tij /K* (if we increase the utilization rate, the cycle time decreases and the production capacity increases). 3. balance the line for each product type j and determine the actual number of stations K j** 4. adjust the line balan cing if needed (ex. keeping the same number of stations for all product types) 5. verify the feasibility of the solution: ∑j Qj * CT j + ∑j SUT j * NB j ≤ H where SUT j = setup time related to product type j, NB j = number of batches of product type j MIXED -MODEL LINES General features of mixed -model lines: - different product types/models can be assembled simultaneously without batching - production rates of product types/models can be adjusted as product demand changes - low inventories are then pos sible In mixed -model lines: - we adjust production rates to demand rates of product types/models, thus limiting inventories - we determine the sequence in which the product types/models have to be launched onto the line - we determine the line balancin g for all product types/models, by balancing within the stations and along the line, so we determine proper allocations of operations to stations by a workload distribution within the stations (station balancing) and along the line (line balancing) - we r educe/eliminate setups as a pre -requisite - we feed the stations: get the right components to each station for the product type/model currently there - we manage production flows when parallel stations are used : the jobs enter the parallel stations followi ng a certain sequence (ex. ABC) that can be different from the sequence in which they exit (ex. BAC), therefore the arrival of the components at the exit of parallel stations must be synchronized with the exit sequence of the jobs; we can us e two types of buffers: ^ resequencing assembly jobs : we put a jobs’ buffer that enables us to recreate the entry sequence of jobs ^ resequencing components (parts) : we put a parts’ buffer that enables us to synchronize the arrival of parts Technical objectives (in a co ntinuous flow line, where the operator can ’t stop the line) are: - minimizing the probability of no completion (station balancing) - minimizing the number of stations, given the cycle time (line balancing) - keeping a constant rate of usage of all compo nents used by the line The steps to follow to balance a mixed -model line are: 1. compute the balancing index within the stations (station balancing): k = index of the station j, w = index of the product type ST = number of stations M = number of product types tjk = mean time of tasks of product type j assigned to station k j = share of quantity dedicated to product type j within the total quantity ( j = Q j/Q, similarly w) The lower the balancing index, the better the balance within each line station. 2. compute the balancing index along the line (line balancing): The lower the balancing index, the better the balance along the line (BI = 0 -> perfect balancing). LESSON 8 - FACTORY PHYSICS Factory physics: - is built on fundamental principles stated as manufacturing laws - is a systematic description of the underlying behaviour of manufacturing systems - is used to support decisions aimed to design new systems, to improve existing systems and to manage manufacturing operations In o ther words, the laws of production logistics are meant to: - express applicable correlations between the production logistics objectives and the corresponding key performances - assist in decision -making, understanding how decisions enable to pursue production logistics objectives of higher delivery capability and reliability with the highest efficiency Example: we have a manufacturing system with a certain weekly demand, an average time to complete the job of no more than one week and no overtime to keep costs low -> we create a graph that links the throughput (units/week) and the throughput time (week): if we increase the throughput, the throughput time increases as the wait ing time increases ; to maintain the throughput time constant, we need overtime. We can use this descriptive model as a support to provide an answer on the feasibility to achieve the target and to make a decision based on a quantitative assessment. TERMI NOLOGY AND SCOPE OF WORK Workstation (station, work -center): a collection of one or more machines or a collection of one or more manual stations that perform identical functions Part: an item that is worked by and moves through the workstations End item ( or finished product) : an item that is sold directly to a customer Routing: a sequence of workstations passed through by a part Order (customer order, replenishment order): a request for a particular part number, in a particular quantity, to be delivered o n a particular date, being it either an actual or forecasted customer/replenishment order Job (work order, production order): a set of physical materials that is transferred through the workstations, along with associated logical information Throughput ( throughput rate, TH): the average output of a production process per unit time Capacity (maximum capacity): the upper limit on the throughput Work in process (WIP): the inventory between start and end points of a product routing Throughput time (TTP): av erage time span a job requires from its release at the beginning of the routing to the end of the routing, when it reaches the end stock point (it can be measure d also for single parts) -> the Little’s Law states that the average number of items in the sy stem is equal to the arrival rate of items (into and out of the system) multiplied by the average time an item spends in the system : WIP = TH * TTP Delivery lead time: the time allotted for the production of a jo