Management Engineering - Operations Management

Completed notes of the course

Complete course

OPERATIONS MANAGEMENT LESSON 1 - OPERATIONS STRATEGY STRATEGY All companies, to survive over time, have to build and keep a specific advantage that differentiate them from competitors. Managing this differentiating advantage is the essence of strategy. We identify three strategic levels: - corporate: we allocate re sources between different markets and products -> where do we compete? - business unit: we define what are the market needs and how to satisfy them -> how do we compete? - functional: we support the company in satisfying market needs According to the tr aditional approach , we define the business unit strategy which develops into the marketing strategy which develops into the operations strategy. Past believes about operations are: - operations are mostly technical - operative aspects are details These statements are no longer true because the world has changed . External relevant factors are: - offer > demand, the offer is much higher than the demand - customisation , the customer asks for a product that fits perfectly with his request - globalisation , the number of suppliers to choose from is very high - high speed of technological development Internal factors influenc ing company’s resource s are : - cultural: the education al level of workers is higher than in the past, everyone can give a contributio n - technological innovation: the many new technologies enable companies to be more competitive - economical: wealth increase , customers are becoming richer - social: authority acknowledgment , companies have to be transparent According to the innovativ e approach, functional strategies cannot be independent one from another, nor they can be the sequential outcome of the business strategy; rather necessity to communicate and interact to define the strategy at business unit and corporate level. The key wor ds are integration and bi-direction . The business model sets the overall purpose and objectives for the operating model, that defines how the business model will be achieved. The realized strategy is composed by: - deliberate strategy: strategy that comes from a rigorous analysis of data - emerging strategy: strategy that comes from unplanned actions and initiatives The sustainable advantage is the outcome of many little good choices. We build up the operations strategy considering four different elements: - top -down perspective: operations strategy should interpret higher -level strategy - bottom -up perspective: operations strategy should learn from day -to-day experience - operations resources: operations strategy should build operations capabilit ies - market requirements: operations strategy should satisfy the organisation’s markets One of the biggest mistakes a business can make is to confuse operations with operational: - operations strategy is longer term - operations strategy is concerned w ith a higher level of analysis - operations strategy involves a greater level of aggregation - operations strategy uses a higher level of abstraction OPERATIONS STRATEGY To define the operations strategy, we first need to understand the customer requir ements: - time , we have many time -related indicators: - time to formulate the offer - time to confirm the order - time to deliver (delivery speed) - delivery reliability (punctuality) - price (cost), the costs the customer has to sustain are: - purchase price - usage cost - maintenance cost - update/upgrade/expansion cost - disposal cost - quality, we have two quality performance: - quality of design , level of quality that the producer wants to offer (specifications) - quality of conformance , level of quality actually offered with respect to specifications - flexibility , we have many types of flexibility: - product, ability to modify the product design to meet market requirements - customization, ability to customize product t o meet market requirements - variety, ability to modify the product mix to meet market requirements - plan, ability to modify your plan to meet market requirements - service , we can offer additional services to customers: - delivery, possibility to choose the supplier (goods availability at the warehouse) - other areas: training, technological improvement, after -sale support Satisfying the customer requirements is difficult and it is not possible to reach the optimum value in all dimensions -> it ’s necessary to align operations and market : we identify the customer needs, we define the market positioning, we analyse the competitors actions and finally we define the desired performances. Then we analyse the levers we have in order to understand wh ich are the achievable performances. The main goal is to make the achievable performances match the desired performances by acting on the levers ; this process is called reconciliation. There are three types of levers: 1. technological/ structural levers - overall production capacity sizing, division and localization - capacity sizing: we can have a low or high capacity - capacity division: we can have a big plant or many small plants - capacity localization: we decide where to locate the plants - stra tegic make or buy : we can do all the activities internally or outsource some of them - technologic process and equipment : we can use automation or not - mechanization/automation grade : we define the optimal level of automation - plant system configurati on : we consider the trade -off between volume production and variety - supply chain configuration : we decide how many suppliers to have and how to relate to them 2. organizational /infrastructural levers - competences needed and their management: we define the needed resources and how to manage them - responsibility allocation: we allocate responsibilit ies among the resources - team vs individuals: we decide whether to work in team or individually - manag ing by objectives or processes: we decide whether to manage by objectives or by processes - functions integration: we can integrate some functions or not - incentive system: we define the incentive system - information flows : we manage the information flows between functions and between levels 3. managerial levers - operations planning and control system - choice of how to meet the demand : we choose the production model (ex. MTS, MTO) - choice of how to realize the product : we choose the production system - supply chain coordination systems: we define the relationship with partners - maintenance managing and realization systems: we can outsource the maintenance or not - continuous improvement systems: we define how many resources and time to dedicate to improvement LESSON 2 - MODELS FOR OPERATIONS STRATEGY PERFORMANCES The performances can be classified in two groups: - order qualifiers (Q) : performance that allows the company to be qualified in the market, but does n’t give the company any competitive advantage . Qualifiers are necessary attributes that a company must possess in order to be a n eligible competitor. Having qualifiers doesn’t increase the company’s profit , but not having them decreases the company’s prof it; in th e latter case the company has order losers (QQ). - order winners (OW) : performance that gives the company a competitive advantage. Order winners are winning attributes that lead customers to purchase. :aving order winners increases the company’s profit. The better the company performs, the higher the number of orders. The entity of the competitive advantage given by an order winner is represented by the slope of the line representing that order winner and depends on the industry in which we co mpete and on the importance of that order winner ( performance ) for customers. Example: two order winners of a smartphone are the quality of the camera and the storage capacity. We should invest on order losers and order winners, but not on order qualifie rs. To define order qualifiers, order losers and order winners we have to: 1. identify a set of representative customers (for example by using the Pareto analysis) 2. identify a set of significant orders ( performances ) by using different tools: - interview: we can interview customers to collect data - ranking : we can ask customers to rank performances - distribute 100 points: we can ask customers to distribute 100 points between performances We have to perform the market segmentation in order to identify the segments to target: - we search and underline differences - each group has homogenous needs - a customer can be part of different groups - we adopt an operations point of view The ranking of performances done by customers has changed ove r the years. The function representing order winners is not always linear; there are other two cases: - saturation behaviour: if a company keeps on improving the performance, the number of orders will increase with a decreasing margin to the poin t in wh ich customers will p erceive no difference . - diverging behaviour: initially a strong improvement of the performance determines a small increase in the orders, but then a small improvement of the performance determines a big increase in the orders. The m ap of performances is a tool that we can use to compare a company with its competitors, but also to study the performances of a company over the years. We put on the horizontal axis a common relative scale and on the vertical axis the performances to evalu ate, then we rate the company and its competitors (or the company over years). This map supports the company in deciding on which performance to invest. We define the priority of interventions based on four elements: 1. importance: how important is the p erformance to customers? 2. distance: how far are we from the minimum required level of performance? 3. competitors: how a re our competitors performing compared to us? 4. difficulties: how much effort and money doe s the intervention requ ire ? LIFECYCLE S CURVE The evolution of the customer requirements can be linked to the product lifecycle, which has four stages: 1. Introduction: the product is first introduced and sales grow slowly. Competitors offering the same product are few or none. Because the customer needs are not clearly defined , the customisation is high or the product design changes frequently. Customers are called innovators and qualifiers are the quality and the product range. 2. Growth: the sales grow rapidly. The number of competitors increases. The product is increasingly standardized in order to supply the rapidly growing market. Customers are called early adopter s and qualifiers are the price and the product range. 3. Maturity: the market is saturated and sales slow down. The number of competitors remains stable. Some types of product domina te while other fade away. Customers represent the bulk of market and qualifiers are the product range and the quality. 4. Decline: the sales decline rapidly. The number of competitors decr eases. The product can be transformed into a commodity. Customers are called laggards and qualifiers are the dependable supply. OPERATIONS STRATEGY EVOLUTION Operations can contribute to the success of the company . Four stages of contribution have been defined: 1. Internally neutral: operations hold the company back and don’t let it succeed; they are a source of competitive disadvantage and must be corrected. 2. Externally neutral: operations make the company as good as competitors; they are not a so urce of competitive advantage but neither of disadvantage. 3. Internally supportive: operations are linked to strategy and make the company the best in the industry; they are a source of competitive advantage. 4. Externally supportive: operations redefin e the industry expectations; they are a source of strong competitive advantage. THE TRADE OFF CURVE (or NO CHANGE CURVE) In literature, the operations manager has two main objectives: - to align the business strategy with the operations strategy - to manage the trade -offs between performance objectives The curve relates two contradictory performances and represents all the ir possible combinations ; its name derives from the fact that the level of competitiveness doesn’t change if we move along the cur ve. The company has to offer a combination of performances that is on the curve; if the company offers one below the curve, it will under produce. The company has to choose between the c ombination s that are on the curve according to its business strategy and the target market to serve . The operations manager can push the frontier through investments, continuous improvements or i nnovations . If he manages to do it, the company gains a competitive advantage and the indu stry expectations are reshaped. The way to move the frontier up and right is to improve both performances: we need to improve cost performance (we get the same level of variety at a lower cost) and variety performance (we get a higher level of variety at the same cost) -> we get more variety at a lower cost. HQ CASE DISCUSSION HQ is a company that produces homeware products. Its biggest customer exited because he perceived products as cheap . In fact the company competes on price, but now wants to produ ce high -quality products. How could have the manager reacted to the exit of their biggest customer (what options)? 1. same product, same market, new customers 2. downsizing 3. follow the customer 4. same product, different market, different customers 5. vertical integration 6. geographical expansion The manager chose the fourth option . Did the manger make the best choice? W e evaluate the other options: 1. if it was possible to find new customers, the company would have done it sooner -> no 2. downs izing is difficult because the company has many fixed assets -> no 3. following the customer means changing the entire supply cain (not exploiting economies of learning ) -> no 5. acquiring the customer is not possible because it is much bigger than the c ompany -> no 6. if the company expanded, it would have to sustain transportation costs and compete on the price -> no  The manager chose the best option . What they did? - they made large investments in new machines , the production capacity remained the same - they introduced a new technology thanks to the bigger size of the new machines, multi -impression mould - they increase d the number of moulds in order to broaden the product range Has been management doing well? Sales are increasing and therefor e revenues are increasing -> good signal Inventor y is increasing more than proportionally to sales -> bad signal Net profit before taxes is increasing -> good signal The customers of the new target market look for quality and therefore they are attentive to indicators such as time to market. The company has to frequently launch new products, that at the end of the season become obsolete or are sold at a very low price. Inventory can be divided into: - physiological inventory (necessary) , it is equal to 10% of revenues (coverage of 5 weeks (5=52*0,1)) - non -physiological inventory, it is equal to inventory minus physiological inventory The non -physiological inventory is probably equal to the value of the unsold products left in stock. It is increasing over the years and this justifies the more than proportional increase in inventory to sales. The non -physiological inventory represents a loss , which is deducted from the net profit.  The system is not under control and the company ha s no idea what will happen in the future. The business strategy of the company is to recover lost revenues by getting (new) different customers. To define the operations strategy, t he operations manager should use the reconciliation model : based on the market requirements, we define the desired performances and the levers we can exploit to achieve them. The desired performances refer to five dimensions and for each target market we identify and rate them: - existing target market - price based competiti on: time reliability (3), cost (3), quality of conformance (1) - new target market - quality based competition: time speed (3), time reliability (1), quality of design (3), quality of conformance (2) , range flexibility (1), plan flexibility (2) The compa ny is not doin g well because the market strategy focuses on effectiveness (quality and variety) while the operations strategy focuses on efficiency (cost). Now we analyse the levers of the company and their impact on the performances: - capacity sizing : the capacity remains the same -> the high saturation has positive impact on cost and negative impact on flexibility and time speed - capacity division: the company reduces the number of machines -> dividing the capacity into fewer machines has positive i mpact on cost and negative impact on flexibility and time - multi -impression mould: the new technology has positive impact on cost and negative impact on flexibility and time speed, moreover it affects the capacity division since it requires big machines - competences: lower fractionation and multi -moulding require few specialized operators -> the high level of specialization has a positive impact on quality of conformance and cost and negative impact on flexibility - responsibilities: there are few re sponsible people who become the bottleneck of the system The company made choices that are coherent with each other, but inconsistent with m arket requirements. The company should have made investments anyway because the old machines don’t meet the new ma rket requirements, but the investments should have been aligned with the market strategy. ONTARIO PACKAGING CASE DISCUSSION Ontario Packaging is a company that produces a wide range of cartons. It wants to replace the two current laminating machines with a new one (the production capacity remains the same). The investment has a payback time of 6,5 years. To achieve the group pa yback time norm of 4 years, the company would need to increase its sales of laminated products by 40% on current levels. Which are the reason why you should approve the investment and the reasons why you shouldn’t? Reasons why you should approve the inv estment: - increase the saturation - solving quality issues Reasons why you shouldn’t approve the investment: - risk on quality conformance - not able to satisfy both customer segments How is the new investment going to change the operations system? Focusing on the traditional customer: - order losers are: quality of design and quality of conformance - qualifiers are: cost and product flexibility - order winners are: plan flexibilit y, time speed and time reliability We analyse the levers of the company and their impact on the performances: - the hig her saturation has positive impact on cost and negative impact on flexibility and time speed - having less people has a positive imp act on cost - bundling the capacity into a single machine determines a higher impact of the setup time, which has negative impact on flexibility (it becomes more impactful to change the production schedule) - the batch size increases due to the higher im pact of the setup time and the higher saturation, it has negative impact on time speed (the waiting time for a product increases) and flexibility  The new investment is going to worsen the order winners. Therefore, it is not possible to sustain the marke t for 4 years with the new configuration proposed , b ut w e must invest because we have quality issue s. Now let’s consider how costs and revenues will really change with the purchase of the new machine: - the difference between current costs and new costs is stable (and positive) as both costs are stable - the purchase of the new machine is a necessary investment to keep the revenues stable ; if we don’t invest, the revenues will decrease -> this point was not considered when computing the payback time  If we consider both impacts of the new investment, the payback time will be shorter than 6,5 years. LESSON 3 - STRATEGIC CAPACITY MANAGEMENT STRATEGIC CAPACITY MANAGEMENT Choosing how much capacity the operations system needs is difficult and should be revised over time. The key factors we have to evaluate when speaking about capacity management are: - timing of change: when we change the overall capacity - magnitude of change : how much we change the overall capacity - attention to transitory phase: how we manage the transitory phase between the current situation and the future situation The key issues we have to face when speaking about capacity management are: - lead time to complete the change - flexibility to change - economie s of scale (the higher the gain, the higher the incentive to increase the capacity) - forecast of demand trend - forecast uncertainty (the higher the uncertainty, the greater the risk of changing the capacity) - competitors behaviour - requested service l evel/customers behaviour Therefore, when determining the overall level of capacity, we have to consider different elements, both internal (related to the operations resources) and external (related to the market requirements). TIMING OF CHANGE When spe aking about capacity management, we can adopt two extreme strategies: 1. leading strategy: we increase the capacity when the level of the demand reaches the current level of capacity (we anticipate the demand) -> the capacity is always higher or equal to the demand 2. lagging strategy: we increase the capacity when the level of demand is higher than the current level of capacity (we follow the demand) -> the capacity is always lower or equal to the demand Pros and cons of the leading strategy are: + al ways spare capacity for opportunities + faster response time (due to spare capacity) + better delivery reliability + lower impact of uncertainty and unforeseen event + lower impact from underestimating demand - higher production unit cost (more fixed cos ts to distribute) - outbound cash flow - higher impact from overestimating demand Pros and cons of the lagging strategy are: + high plant utilisation + low production costs (less fixed costs to distribute) + lower impact from overestimating demand - longer response rate (due to no spare capacity) - lower delivery reliability - higher impact from underestimating demand The choice of the strategy depends on the industry where we compete and our positioning in the industry. But we can also adopt inter mediate strategies: 1. smoothing strategy: we use the spare capacity of one period to produce inventory which can be used to supply the under capacity period (we try to be aligned with the demand rate) -> the capacity is alternatively higher and lower tha n the demand . We can use this strategy if we produce inventories that can be used in the mid -term ; otherwise we have to adopt other strategies, such as the following ones. 2. filling products strategy: when the capacity is higher than demand, we use the s pare capacity to produce filling products or we active ly outsource the production . It is useful to identify which products sacrifice when the capacity is lower than demand. 3. outsourcing strategy: we don’t have spare capacity and when the capacity is lower than demand we passively outsource the production. The lifecycle of the product affects the strategy capacity management: - in the introduction phase , companies usually adopt the leading strategy to gain market share - in the growth phase, companies usually adopt an intermediate strategy looking for efficiency - in the maturity phase, companies usually adopt the lagging strategy as sales starts to decline The demand we compare with the c apacity is forecasted and therefore suffers from uncertainty. The demand can assume a range of values between the upper forecast and lower forecast. The level of uncertainty of the forecast is not stable, it typically increases as the time increases. When we choose the strategy, the timing of change and the magnitude of change, we have to consider this level of uncertainty ; examples are: - if the level of uncertainty is low, you can adopt the leading strategy - if the change is big, the impact of uncertain ty becomes very relevant - if you don’t want to risk, you can set the investment with a lower bound of uncertainty, otherwise with an upper bound of uncertainty MAGNITUDE OF CHANGE In case of an investment proposal, it is important to perform a scenario analysis: we evaluate possible scenarios that could take place in the future. It must be done: - to evaluate the risk of the investment - to identify possible constraints/opportunities - to prepare countermeasure - to limit the impact of unpredictable events - to prepare a faster, more efficient and more effective response Each choice is evaluated on the basis of two elements: the economic impact and the risk , that depends on the uncertainty and the variability. All the choices have a cost ( even not i nvesting) -> no risk no gain. Elements to consider when deciding the size of the capacity increase are: - technological constraints - economies of scale: the bigger the increase, the higher the economies of scale (but also the risk) - demand uncertainty : the bigger the increase, the higher the impact of the uncertainty - financing availability: we need money to increase the capacity - over/underutilisation costs - outsourcing possibility LESSON 4 - INTRODUCTION TO SERVICE AND SERVICE PROCESSES INTRODUCTION TO SERVICE The service industry is growing and the main causes of this growth are: - social and demographic trend: evolution of needs, welfare, entertainment - ICT: services to distribution and communication - globalisation: transport and t ourism, services for industrial companies - outsourcing: services for industrial companies According to the Maslow’s hierarchy of needs , people are motivated by five core needs which can be organized into a hierarchy. From the bottom of the pyramid, the needs are: physiological needs, safety needs, love and belonging, esteem, self -actualization. When a lower need is fulfilled, a higher need emerges . Each need requires the execution of specific processes to be satisfied , therefore operations is crucial . The service industry is much more complex t han the manufacturing industry and this derives from the differen ce between a service and a physical product ; a service is characterized by: - intangibility : the service is intangible (creative advertising, importance of reputation) - customer participation: the customer is involved in the service process (co -production) - simultaneity: the production and the consumption of the service are simultaneous - perishability: the service can not be stored , we need to match supply with demand - heterogeneity: the result of the service process is variable due to customer participation SERVICE PROCESSES The process is one key element to define a service delivery system. It can be classified b ased on: 1. interaction with customer (visible or not visible to customer): operations of a company can be : - front -office focused: the experience part is very important, the outcome is taken for granted and is not differentiated ; customer management is essential (ex. cinema) - back -office focused: the experience part is less important, the outcome is essential (ex. courier) In general, the level of front -office and back -office is set by the sector. However, the operations manager can use levers, such as centralization: if you are able to decouple back -office activities, then you can centralize them. Pros and cons of centralization are: + efficiency , thanks to higher volume ( higher saturation of resources ) + specialization of activities, thanks to higher volume + possibility to follow personal attitudes + less volume variability - possible gap with the front -office - longer lead times and greater rigidity - overlapping activities (some activities are executed at the same time) 2. volume vs variety offered : according to the volume -variety matrix, there are four types of service: - mass services: low variety (standardised products) and high volume (ex. McDonald’s) - standardised processes - short interaction with customer (back -office oriented ) - focus on productivity and conformance - automation/informative systems - competences embedded in the system - competitive advantage gained through process innovation - mass service s shops: developm ent of mass services towards a broader service offer - request of increasing variety without losing control over the processes - increasing of the discretion degree (more differentiated) - front -office people need to develop the ability to understand customer needs - professional services shops: development of professional services as dimension increases - request of increasing efficiency without losing customization - request of know ledge sharing (embedded in the system) - development of a “house style” - decreasing of the discretion degree (more standardised) - creation of semi -professional roles - development of roles dedicated to interactions with customer - professional serv ices: high variety (differentiated products) and low volume (ex. consultancy) - not well defined/standardised processes - long interaction with customer (front -office oriented) - focus on providing solutions - people’s competences are a critical ass et - competitive advantage gained through product innovation To better understand the variety degree that the process needs to manage, we divide service requests in: - runners: requests that always need the same operations/activities; often foreseeable and in remarkable volumes; opportunity for automation and process review . - repeaters: requests that refer to known activities, but clustered in a different way; not so much foreseeable and in medium -low volume; they are expected events, but not frequent . - strangers: requests that need the design of new activities; often a bit foreseeable; not expected events. Example: car service -> oil change; gearbox repair; product recall A company can decide where to focus using the key decisional area matrix: o n the vertical axis we have the types of service request and on the horizontal axis the customer involvement. There are four quadrants: - service projects: operations should focus on front -office and back -office - service partnership: operations should fo cus on customer and front -office - service factory: operations should focus on back -office - diy service: operations should focus on customer, front -office and back -office Variability is the gap between the actual value and the average value. Uncertainty is the gap between the actual value and the expected value. Not explained variability causes uncertainty -> uncertainty can be reduced by explaining variability: thus it is possible to reduce uncertainty and simply have variability. There are two types of variability: - determined by the company -> we can remove it (if not possible, we isolate it) - outer variability -> we can limit its effect The Pareto principle is valid: 20% of the products/services generates 80% of variability. In a service system, variability has to be managed at the front -office. The greater the variability, the greater the competences and discretion needed by the front -office. A system based on command and control is very inefficient (and often even ineffectiv e) in managing variability. 3. processing: to design the service process we have to consider people, objects, information LESSON 5 - SERVICE CONCEPT SERVICE CONCEPT The service concept aims at aligning the system towards the inside and towards the o utside. Aligning the goal is crucial because the process is not totally controlled. The service concept aims at reducing the difference between how the company would like the service to be perceived from clients, employees and stakeholders and how client s, employees and stakeholders see the service. The service concept is made of three parts: 1. organising idea, it is the essence of the service brought to or used by the customer It is a wonderful alignment tool for employees and for customers, a guideli ne for all choices, a continuous support of the competitive differentiation of the service offered. 2. service provided, it is composed by: - the processes to create and deliver the service - the output, the result seen from the company’s perspective - the service delivery system, the environment where the service is delivered 3. service received, it is composed by: - the outcome, the result seen from the customer’s perspective - the experience, customer’s perception of the service delivery In design ing the experience we have to consider different elements: - customisation of the process - response speed (of the delivery system) - employees flexibility - intimacy with the customer - accessibility of the personnel - perception of being valuable - curtesy and competence SERVICE PROFIT CHAIN According to the service profit chain, profit and revenue growth are stimulated by the customer loyalty, which is a direct result of the customer satisfaction. The customer satisfaction depends on the valu e of the services provided to customers. This value is created by satisfied, loyal and productive employees . Employee satisfaction results primarily from high -quality support services and policies that enable employees to deliver results to customers. Overall, to increase the customer loyalty the company should focus on creating satisfied customers by providing high value. Understanding what customers value is difficult and to deal with it the company can train the frontline employees to understand how t o translate customer feedback into useful input. CUSTOMER VALUE The customer value can be computed as: customer value = outcome + experience It can be compared to: price (for the customer) + acquisition cost Many measures of customer retention and cus tomer loyalty are focused on past analysis and not on new elements that will ensure customers’ future loyalty. Surprisingly there is more attention on finding new customers than keeping those already engaged, as if customers are inalienable.  Keeping cu stomers loyal to you should be your priority. The customer satisfaction depends on the perceptions of the service received compared to the expectations : - if the perceptions are higher than the expectations, the customer is delighted . - if the percep tions are equal to the expectations, the customer is satisfied. - if the perceptions are lower than the expectations, the customer is dissatisfied. The expectations derive from the reputation of the company and the price. The mismatch between perceptions and expectations is the sum of two gaps: - gap 1 : there is a gap between the expectations on the service and the service actually received - gap 2: there is a gap between the value of the service received and the percepti ons on the service  There is a zone of tolerance where expectations are aligned with perceptions, which is acceptable for companies and makes customers satisfied. SHOULDICE HOSPITAL CASE DISCUSSION The Shouldice hospital is a private clinic specialized in abdominal hernias , that operates in Canada . They are well known and have a long queue of customers. In fact, the owner wants to extend the business. Actually, he has already tried to do it by working also on Saturday. At the same time, they are facing the inc rease of small competitors. What is the value proposition of the company ? They do an excellent work, take care of patients, provide a quality service at a lower price. They build intimacy among patients and among the patient and staff. Everything is explained in one visit, the recovery is fast. The salary is higher and the working time is low . They are specialized in a single operation. The service concept of the company is made of three parts: 1. organising idea: solving hernia, but with points of differentiation: the customer feels on vacation, he has no worries and the service is accessible to everybody 2. service provide d (the output ): social interaction , low recovery time, self -recovery 3. service perceived , which is composed by : - the outcome: cheap service, quality of conformance and quality of design - the experience: intimacy, confidence , on vacation To push social interaction, t he company uses the following levers : - choice of having double rooms instead of single ones (structural decision) - same dining room for patients and staff (structural decision) - same lunch time for patients and staff (managerial decision) - choice of putting similar people in a room (managerial decision) To assure a low recovery time, the c ompany uses the following levers: - choice of curing few types of customers, with good health (managerial decision) - specialization in certain types of hernia (organizational decision) - standardized procedure (organizational decision) To enable a self -recovery, they have created an environment where it is easier for patients to rehabilitate on their own (special stairs, carpets, moquette) (structural decision). It is possible to keep the service c heap because there are fewer nurses than in the public ho spital and the customer does his self -diagnosis. Short and clear working time and high salary allow to achieve a good working environment. They did this to balance the fact that the job isn’t that motivating (it’s the same everyday) and it is difficult to make career. How can the company increase its business? They tried to expand the business by working also on Saturday, but it didn’t work because there were two groups in the staff (a group working on Saturday and the other not) - > the staff is not happy. The owner is considering the option of opening a new facility in USA. The challenges of this option are: - loss of control since the new facility is far from the first one - it is a large investment and it could increase the cost of the service - doctors have to be split among the two facilities The real difficulty is that the company has to replicate the entire system, also the soft aspects that make the company successful. The soft aspects are all those characteristics built over time, they are the result of many small improvements performed in a long time. LESSON 6 .1 - QUEUE MANAGEMENT QUEUEING SYSTEM A queueing system is formed by one or more customers waiting to be served by one or more servers. Customers’ examples are: - people waiting at the cashier ’s desk for a bank operation -> corresponding server = cashier - pieces waiting to be processed by a lathe in a job shop -> corresponding server = job shop lathes - container waiting to be loaded on chassis -> corresponding server = overhead traveling cran es We can’t avoid queues, but we can manage our system to reduce the waiting time in queues. QUEUE FORMATION Queues form due to an imperfect balance between demand rate and service rate: - structural imbalances: they characterize a system in which the service rate is lower than the arrival rate - incidental imbalances: they depend on: - the variability : difference between the actual and the average value of the arrival time - the uncertainty : difference between the actual and the expected value of the arrival time Usually, the tendency is to consider average values only; however, there is always a variability that needs to be considered and managed. Forecast and historical data help in mitigating possible queue impacts on system performances, but they are not the only solution. The structural condition of a system is that the average rate of customers’ arrival into the system is lower than the average of the server’s serving rate. -> Queues are a result, a symptom. Queues are particularly critical in the service companies because if a customer is not satisfied because he had to wait in line for a long time, he might tell other potential customers (word of mouth). QUEUING SYSTEM MODELING The customer can assume di fferent behaviours : - rejecting: when a customer is rejected by the system because he doesn’t respect acceptance requirements - balking: when a customer decides not to enter a queue because it’s already too long - reneging: when a customer already in queu e gives up the service and goes away without being served - jockeying: when a customer tries to trick the rules to get advantage, for example shifting from one queue to another to shorten the waiting The queuing system parts are: ^ calling population: customer population is the input source; it can be: - finite: if the potential number of new customers for the system is significantly affected by the number of customer already in the system - infinite: if the number of customers i n the system does not affect the demand rate for the service made by new customers ^ arrival process: the arrival process describes how customers show up; it’s described by the distribution of interarrival time, which represents the time interval occurrin g between two consecutive arrivals (very often the interarrival time distribution is well represented by a negative exponential distribution) ^ service process: the service process describes ho w every server delivers the service, in particular it defines its duration; it’s described by the distribution of service time (often the service time distribution is well represented by a negative exponential distribution) ^ queue configuration : there are different queue configurations: - single queue: customers form a single line and once they reach the front of the line, they go to the available server; the number of servers can be equal to or greater than one Pros: the first customer arrived is the first one served (FCFS), it avoids the jockeying problem; it minimizes the average waiting time; reneging actions are less frequent as the waiting time is lower Cons: it could “scare” the customer (and therefore cause balking actions) - multiple queue: each server has its own queue and customers choose which one to e nter Pros: it allows to diversify the service and the work; customers can choose a servicer of his liking ; balking actions are less frequent Cons: with the same number of servers, the average waiting time is higher than in single queue - take a number: it’s a variation of the single queue; the customer takes a number and waits to be called by a server; the number of servers can be equal or greater than one Pros: the first customer arrived is the first one served (FCFS); it avoids anxiety related to the thought that another queue could be faster; it minimizes the average waiting time; customers have the possibility to relax and dedicate themselves to other things during the wait Cons: an absent -minded customer could risk losing its turn; it could “scare” the customer - infinite servers: the customer is his own server (self -service) Some elements for the choice of the queue configuration are: - service customization degree: if we want to increase the customization, the multiple queue is the best - service time variability: if the variability and unpredictability are high, the single queue is the best As regards the queue capacity we can have: - limited queue : the number of customers in line plus the ones who are being served is limited; customers who arr ive after the capacity is overfilled are rejected - unlimited queue: the number of customers in line plus the ones who are being served is unlimited ^ queue discipline: it is the rule or set of rules related to the order in which customers are served, it’ s often strictly related to the queue configuration. It can be: - static: the pertaining order among the customers in queue doesn’t change in time and/or at the changing of the system conditions - dynamic: the pertaining order among the customers in queu e can change in time and/or at the changing of the system conditions QUEUE SYSTEM ANALYSIS TOOLS There are three methods that we can use to analyse the queu eing system: 1. deterministic analysis: - pros: it is simple and intuitive to apply - cons: it d oesn’t consider transitory, arrival time variability and service time variability; queue is considered as an on -off aspect 2. queueing theory: - pros: it considers inter -arrival time variability and service time variability; it allows to calculate a set of significant variables - cons: it doesn’t consider transitory, mathematic analysis can become very complex or even intractable in case of complex waiting lines; it doesn’t consider customers’ complex behaviours 3. simulation: - pros: it considers trans itory; it is very flexible: - for the type of systems that can be shaped - for the type of data that can be obtained - economic and user friendly software are always more widespread - cons: it is time consuming; skills are needed to design, production and analysis The assumptions of the queueing theory are: - stationary process: this assumption could be very restrictive -> usually the arrival process isn’t a static process, it changes during the day. A solution is to make more analysis considering different time slots . - peculiar customers behaviours are not expected (no balking, reneging or jockeying) - true only for certain inter -arrival time and service time distributions Kendall’s codification turns out to be A/B/c/K/m/Z: - A identifies cust omers' arrival distribution - B identifies service time distribution - c identifies number of servers - K identifies queue capacity (buffer by default: endless) - m identifies population dimension (by default: endless) - Z identifies service discipline (by default: FIFO) Usually stop at the first 3 components (ex. M/G/1), in detail: - M suggests a negative exponential distribution (Markovian) - G suggests a Gaussian distribution Some examples of model are: - standard M/M/1, standard M/M/c -> most common - finite queue M/M/1 (M/M/1/K), finite queue M/M/c (M/M/c/K) -> particularly useful in estimating lost sales due to an inadequate waiting area or to an excessive long queue - general self -service M/G/ ∞ -> the total number of customers in the process varie s due to arrival time and service time variability. This model is also useful to shape rounding situations where you rarely have to wait The relevant parameters are:  = arrival rate (customers per time unit)  = service rate (customers per time unit) n = number of customers in the system (waiting + serving) Lq = average number of customers in line Wq = average waiting time in line by customers Ls = average number of customers in the system Ws = average waiting time in the system by customers /(*s) = system utilization rate -> it’s the coefficient that describes the utilization degree of the system, where s represents the number of servers Pn = probability that there are n customers in the system Po = probability that there are no costumers in t he system In a M/M/1 queueing system : P(n ≥ k) = (/)k = k Po = 1 -  Pn = Po*n Ls = /(-) Lq = /(-) Ws = 1/(-) Wq = /(-) Example s: slide 48 -54 + 55 -58 The waiting time doesn’t decrease in a linear way with the increase of the number of counters. The waiting time doesn’t grow in a linear way with the increase of the system saturation rate . The queueing theory is addressed because of: - system sizing a nd designing - evaluation of costs and gains to improve the service - queue length (long line -> poor service) - average waiting time in line (long wait -> poor service) - average waiting time in the system (long wait -> poor service) - system utili zation rate Two achievable goals when sizing a system are: 1. service quality principles - average waiting time of the customer (banks, restaurants): E(W*) <  - maximum waiting time (public services): P(W* > w) ≤  2. cost principles - minimization of the sum of the wait cost and the service cost (trade -off) In case of no variability, the queue length increases exponentially as utilization increases. Process variability results in both waiting and resource underutilization : if the v ariability is high we have moments in which: - the utilization is high, but the waiting time is long - the waiting time is short, but the utilization is low In order to move from the point X (where the waiting time is too long), we can act in two ways: - we reduce the utilization and we move to point Y (ex. by increasing the number of servers) - we reduce the system variability and we move to point Z (ex. by introducing a booking system) The point Z represents the best situation: high utilisation and s hort waiting time due to reduced variability. Managers modify the queueing system in order to get better performances. The set of levers the manager takes to manage queues are: - levers and c ountermeasures offer -side - configuration changes - add, re move or move resources - decrease service time (technology utilization, training) - increase resources flexibility - decrease service time variability - levers and countermeasures demand -side - decrease uncertainty level (booking, improve forecasts) - decrease variance (incentives for non -rush hours moments, booking) - levers and countermeasures related to the psychology of waiting There are two important laws in the customer service: 1. Satisfaction = perception - expectation -> perception is mo re important than reality 2. First impression is the most important Measuring the customer satisfaction is important to understand where the critical point of the system is. The impact of one dissatisfied customer is much more than the impact of one sat isfied customer. We have to know that: - the unoccupied time seems longer than occupied time -> we can distract and entertain with small related or unrelated activities (ex. magazines at the doctor) - pre -process wait seems longer than in -process wait -> we should take care of the customers as soon as possible to make him/her “in -process” (ex. tracing package with Amazon) - anxiety makes the wait seem longer -> we can communicate frequently the waiting time to the customer We also have to consider that: - an unfair wait seems longer than a fair one - the more the service is valuable, the more the customer is willing to wait - waiting alone seems longer than waiting in a group - a customer exposed to an exaggerated wait will be a more difficult one to se rve or an ex -customer Therefore be careful to all these factors and make sure to organize some actions to avoid that the customer senses the supplied service in a negative way. LESSON 6.2 - QUEUE MANAGEMENT In a M/M/c queueing system: P(n ≥ k) = ( /)k = k Po = 1 -  Pn = P o*n Lq = f( /; c) -> from table  Ls = L q +  Wq = L q/ Ws = L q/ + 1/  Example: slide 4 -8 COMPLEX SYSTEM A complex system is the composition of more elementary sub -systems that interact among themselves (mapping the system and identifying every step of the process). Very often in a complex system different types of customers exist (we must know in advance eve ry type of customer in the system). The system expected throughput time is the representation of an average customer and not of the specific one. The throughput time varies according to the variation of the workload and the variation of the input. Example : slide 9 -15 NODE BALANCING Everything that enters in the node is equal to everything that comes out from it. =n the system there isn’t any flow loss. Example: slide 17 -24 M/M/1 SYSTEM WITH ASSIGNED PRIORITY TO A CUSTOMER CLASS =t’s a system formed by just one server who is able to serve two classes of customers (1 and 2) that have a different service admission: customers of class 1 have a service admission priority higher than customers of class 2. The following two cases will b e analysed: - preemptive priority: if a customer of class 1 enters the system while a customer of class 2 is being served, the service to the customer of class 2 is interrupted in order to give immediate precedence to customer of class 1. The system throu ghput time of the two classes of customers can be computed as: E(S 1) = (1/ )/(1 -1); E(S 2) = (1/ )/[(1 -1)*(1 -1-2)] - non -preemptive priority: if a customer of class 1 enters the system while a customer of class 2 is being served, the service to the customer of class 2 is completed and then the customer of class 1 is served. The system throughput time of the two classes of customers can be computed as: E(S 1) = [(1+ 2)/]/(1 -1); E(S 2) = [(1 -1*(1 -1-2))/ ]/[(1 -1)*(1 -1-2)] The alteration of the priority logic doesn’t change the system expected throughput time. :owever, in some cases it’s essential to increase the cus tomer satisfaction degree and to improve customer service. EXERCISE 1 + 2 + 3 + 4 LESSON 7 - YIELD MANAGEMENT The fundamental responsibility of operations management is to provide the capability to satisfy current and future demand. It has to manage the trade -off between customer service and cost. How can we cope with mismatches between the demand and the capability of the service delivery system? How can we manage variability in our system? Yield management answers these questions. YIELD MANAGEMENT In 1978, Airline Deregulation Act put a price liberalization in the airline field. Companies had to face the uprising of new companies with lower prices. In this background, American airlines had the idea to saturate the seats capacit y offering at lower cost some of the unoccupied seats. It has been necessary to determine seats best allocation between economy and business classes to avoid the cannibalisation effect (having customers willing to pay high prices who buy low costs seats). To avoid rate cannibalisation, they decided to offer service diversification. The biggest difficulty was to allocate capacity to the different classes offered per flight, in order to avoid under/overestimation of airplane tickets. Revenues have been incre ased of 1,5 billion dollars in 3 years. Yield management exploits information about customers’ behaviour obtained by operations department while delivering the service. This allows to improve the competitiveness of the company. The company acts in an in tegrated way on demand and capacity in order to sell the right capacity, to the right customer, at the right time and at the right price so to maximise the profit. Yield management systems refer to strategies and tactics used by a certain number of compan ies to manage the allocation of their capacity to different rate classes to maximise their profitability. The main goal is to maximise the capacity utilisation rate to reach profitability as close as possible to the maximum achievable one. Therefore, yield management is a systematic approach to maximise profitability by offering price differentiation to potent ial clients. Application areas are constantly increasing. Companies adopting yield management operate in transport, entertainment, hotels and restaurants. They have some common characteristics: - fixed capacity - uncertain demand - low marginal sales c osts and high fixed costs - ability to segment markets - limited capacity for different market segments - perishable inventory - product booked/sold in advance - high capacity change cost The possible configurations of the process are: 1. booking in ad vance, purchasing (payment) and then service delivery 2. booking in advance, concurrent purchasing and service delivery 3. concurrent booking and purchasing and then service delivery 4. concurrent booking, purchasing and service delivery YIELD MANAGEM ENT TOOLS The yield management tools are: 1. capacity allocation: we decide how much capacity to allocate to each segment a) price policies - ideally, customers are clustered in classes according to their willingness to spend; we want to maximize the pr ofit so we define the objective function: max ∑j Mj * X j , subject to: X j ≤ Nj , ∑j Xj ≤ C where X j = number of tickets allocated to class j, M j = unit margin for a sale to class j, N j = number of customers belonging to class j, C = overall capacity - in reality, customers have different behaviours; the demand from the cheapest classes usually appears before the demand from the expensive classes -> the cheapest classes buy in advance to save money while the expensive classes buy just before the event; ther efore, the cheapest classes cannibalize the expensive classes. We need to allocate the right capacity to full price customers and the remaining capacity to low costs customers -> our target customers are the full price customers. b) demand forecast We for ecast the demand of each class of customers (from now on we consider of having two classes). c) protection policies We first receive the demand of discounted price customers, we accept or reject the request from discounted price customers and then we rec eive the demand of full price customers. So, we first define the protection level (S 1), which is the protected capacity allocated to full price customers, and then the capacity allocated to discounted price customers (A 2 = C - S1). To define the protection level we can use two different methods: - marginal analysis: we define three variables: X1 = demand of a full price unit Cu = cost to underestimate the demand of a full price unit (the demand is higher than expected) Co = cost t o overestimate the demand of a full price unit (the demand is lower than expected) We define the objective function (balance between underestimate cost and overestimate cost): P(X 1 ≥ S1)*C u ≥ P(X 1 < S1)*C o -> [1 - P(X 1 < S1)]*C u ≥ P(X 1 < S1)*C o -> P(X 1 < S1) ≤ C u/(C u + C o) =  -> we find  from the table of the standard normal distribution -> S 1 =  + * The higher the protection level, the lower the underestimation cost and the higher the overestimation cost. The marginal analysis can be used not only for service industry, but also for manufacturing industry. The protection level is used for all cases when you want to protect some capacity for more profitable customers which will probably arrive later on (ex. d resses of a new collection, newspapers to resell, bread to prepare). Examples: slide 16 -18 + 26 -29 (presentation 7.2) - heuristic expected marginal seat revenues (EMSR) : we define six variables: n = number of available classes fi = unit revenue associat ed with class i (f 1 ≥ f 2 ≥ … ≥ f n) i = average demand for class i i2 = variance in demand for class i i = level of protection for class i and more expensive classes Di = available demand to pay class i, or more expensive classes We fix the protection level for each class: Example: slide 9 -11 (presentation 7.3) 2. overbooking: we ca n use overbooking in two situations: - overbooking, due to the fact that not all bookings become a sale - overbooking of sold tickets over capacity (overselling), due to the fact that not everyone who bought the ticket will use the service (no -show s) Therefore, the overbooking can be applied to two decision points: - overbooking on the number of reservation s: we accept a number of reservations higher than the actual capacity in order to protect ourselves from the reservations that do not turn into purchases - overbooking on the sales: we sell a number of tickets that is bigger than the actual capacity in or der to protect from the no -show effect The application of the overbooking depends on the process type -> w e consider the third one ( concurrent booking and purchasing and then service delivery ). There are different methods that we can use for overbooking: - analysis of the mean values: we set the maximum number of units in overbooking: BL = C/p where BL = booking limit, C = capacity, p = probability that the person who has booked the ticket, then buys it It is easy to understand and to implement, but it do esn’t consider the costs. - fixed service level: we set the level of service to be provided to customers: P(X > C) ≤ l (the probability that the number of units in overbooking is greater than the number of no -show is lower than or equal to l) -> 1 - P(X ≤ C) ≤ l -> P(X ≤ C) ≥ 1 - l - analysis with calculation paper - marginal analysis: we continue to accept reservations until the expected margin on the last booking accepted is greater than or equal to the expected loss on the last booking Ovb = number of units in overbooking; NS = number of no -show Cu = cost to underestimate no -show (no -show is higher than expected) Co = cost to overestimate no -show (no -show is lower than expected) We define the objective function (balance between underestimate cost and overestimate cost): P(Ovb ≤ NS)*C u ≥ P(Ovb > NS)*C o -> P(Ovb ≤ NS)*C u ≥ [1 - P(Ovb ≤ NS)]*C o -> P(Ovb ≤ NS) ≥ Co/(C u + C o) =  -> we find Ovb from the table - dynamic method There are difficulties in the implementation of YM systems: - customers reaction -> difficulty in understanding, attempts to fool the system - conflicts of objectives among different business areas - cost/time of implementation -> cost of the information system, time/cost/difficulties in data collection EXERCISE 1 + 2 + 3 LESSON 8 - COPYING WITH VARIABILITY AND UNCERTAINTY Variabili